Hugging Face's logo

Datasets:
YFanwang
/
tinyllava

Dataset card Files
xet
Community
YFanwang commited on
Commit
5143b46
·
verified ·
1 Parent(s): 789392e

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation_20251018_062713.log +11 -0
  2. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation_20251018_062714.log +11 -0
  3. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation_20251018_062715.log +11 -0
  4. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_0.5_2e-1_connector-3.0_0.5_2e-1_ablation_20251018_062715.log +11 -0
  5. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_2.3_2e-1_connector-3.0_2.3_2e-1_ablation_20251018_062716.log +11 -0
  6. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.5_2e-1_connector-5.0_2.5_2e-1_ablation_20251018_062716.log +11 -0
  7. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.7_2e-1_connector-5.0_2.7_2e-1_ablation_20251018_062717.log +11 -0
  8. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.9_2e-1_connector-5.0_2.9_2e-1_ablation_20251018_062718.log +11 -0
  9. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_0.9_2e-1_connector-7.0_0.9_2e-1_ablation_20251018_062718.log +11 -0
  10. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.1_2e-1_connector-7.0_1.1_2e-1_ablation_20251018_062719.log +11 -0
  11. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.3_2e-1_connector-7.0_1.3_2e-1_ablation_20251018_062719.log +11 -0
  12. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.5_2e-1_connector-7.0_1.5_2e-1_ablation_20251018_062720.log +11 -0
  13. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.7_2e-1_connector-7.0_1.7_2e-1_ablation_20251018_062720.log +11 -0
  14. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.9_2e-1_connector-7.0_1.9_2e-1_ablation_20251018_062721.log +11 -0
  15. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.1_2e-1_connector-7.0_2.1_2e-1_ablation_20251018_062721.log +11 -0
  16. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.3_2e-1_connector-7.0_2.3_2e-1_ablation_20251018_062722.log +11 -0
  17. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.5_2e-1_connector-7.0_2.5_2e-1_ablation_20251018_062722.log +11 -0
  18. ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.7_2e-1_connector-7.0_2.7_2e-1_ablation_20251018_062723.log +5 -0
  19. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/added_tokens.json +24 -0
  20. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/config.json +88 -0
  21. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/generation_config.json +7 -0
  22. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/log.txt +2 -0
  23. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/merges.txt +0 -0
  24. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/model.safetensors +3 -0
  25. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/runs/Oct17_08-17-13_ywang29-vrdb-test1-worker-0/events.out.tfevents.1760689110.ywang29-vrdb-test1-worker-0 +3 -0
  26. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/special_tokens_map.json +32 -0
  27. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/tokenizer_config.json +208 -0
  28. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/trainer_state.json +3670 -0
  29. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/training_args.bin +3 -0
  30. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/vocab.json +0 -0
  31. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/added_tokens.json +24 -0
  32. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/config.json +88 -0
  33. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/generation_config.json +7 -0
  34. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/log.txt +2 -0
  35. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/merges.txt +0 -0
  36. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/model.safetensors +3 -0
  37. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/runs/Oct17_08-51-11_ywang29-vrdb-test1-worker-0/events.out.tfevents.1760691142.ywang29-vrdb-test1-worker-0 +3 -0
  38. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/special_tokens_map.json +32 -0
  39. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/tokenizer_config.json +208 -0
  40. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/trainer_state.json +3670 -0
  41. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/training_args.bin +3 -0
  42. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/vocab.json +0 -0
  43. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/added_tokens.json +24 -0
  44. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/config.json +88 -0
  45. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/generation_config.json +7 -0
  46. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/log.txt +2 -0
  47. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/merges.txt +0 -0
  48. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/model.safetensors +3 -0
  49. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/runs/Oct17_09-25-10_ywang29-vrdb-test1-worker-0/events.out.tfevents.1760694106.ywang29-vrdb-test1-worker-0 +3 -0
  50. ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/special_tokens_map.json +32 -0
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation_20251018_062713.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation_20251018_062713.log
3
+ Timestamp: 2025-10-18 06:27:13
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation_20251018_062713.log
10
+ Timestamp: 2025-10-18 06:27:14
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation_20251018_062714.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation_20251018_062714.log
3
+ Timestamp: 2025-10-18 06:27:14
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation_20251018_062714.log
10
+ Timestamp: 2025-10-18 06:27:15
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation_20251018_062715.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation_20251018_062715.log
3
+ Timestamp: 2025-10-18 06:27:15
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation_20251018_062715.log
10
+ Timestamp: 2025-10-18 06:27:15
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_0.5_2e-1_connector-3.0_0.5_2e-1_ablation_20251018_062715.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_0.5_2e-1_connector-3.0_0.5_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_0.5_2e-1_connector-3.0_0.5_2e-1_ablation_20251018_062715.log
3
+ Timestamp: 2025-10-18 06:27:15
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_0.5_2e-1_connector-3.0_0.5_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_0.5_2e-1_connector-3.0_0.5_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_0.5_2e-1_connector-3.0_0.5_2e-1_ablation_20251018_062715.log
10
+ Timestamp: 2025-10-18 06:27:16
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_2.3_2e-1_connector-3.0_2.3_2e-1_ablation_20251018_062716.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_2.3_2e-1_connector-3.0_2.3_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_2.3_2e-1_connector-3.0_2.3_2e-1_ablation_20251018_062716.log
3
+ Timestamp: 2025-10-18 06:27:16
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_2.3_2e-1_connector-3.0_2.3_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_2.3_2e-1_connector-3.0_2.3_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-3.0_2.3_2e-1_connector-3.0_2.3_2e-1_ablation_20251018_062716.log
10
+ Timestamp: 2025-10-18 06:27:16
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.5_2e-1_connector-5.0_2.5_2e-1_ablation_20251018_062716.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.5_2e-1_connector-5.0_2.5_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.5_2e-1_connector-5.0_2.5_2e-1_ablation_20251018_062716.log
3
+ Timestamp: 2025-10-18 06:27:16
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.5_2e-1_connector-5.0_2.5_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.5_2e-1_connector-5.0_2.5_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.5_2e-1_connector-5.0_2.5_2e-1_ablation_20251018_062716.log
10
+ Timestamp: 2025-10-18 06:27:17
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.7_2e-1_connector-5.0_2.7_2e-1_ablation_20251018_062717.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.7_2e-1_connector-5.0_2.7_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.7_2e-1_connector-5.0_2.7_2e-1_ablation_20251018_062717.log
3
+ Timestamp: 2025-10-18 06:27:17
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.7_2e-1_connector-5.0_2.7_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.7_2e-1_connector-5.0_2.7_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.7_2e-1_connector-5.0_2.7_2e-1_ablation_20251018_062717.log
10
+ Timestamp: 2025-10-18 06:27:17
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.9_2e-1_connector-5.0_2.9_2e-1_ablation_20251018_062718.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.9_2e-1_connector-5.0_2.9_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.9_2e-1_connector-5.0_2.9_2e-1_ablation_20251018_062718.log
3
+ Timestamp: 2025-10-18 06:27:18
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.9_2e-1_connector-5.0_2.9_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.9_2e-1_connector-5.0_2.9_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-5.0_2.9_2e-1_connector-5.0_2.9_2e-1_ablation_20251018_062718.log
10
+ Timestamp: 2025-10-18 06:27:18
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_0.9_2e-1_connector-7.0_0.9_2e-1_ablation_20251018_062718.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_0.9_2e-1_connector-7.0_0.9_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_0.9_2e-1_connector-7.0_0.9_2e-1_ablation_20251018_062718.log
3
+ Timestamp: 2025-10-18 06:27:18
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_0.9_2e-1_connector-7.0_0.9_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_0.9_2e-1_connector-7.0_0.9_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_0.9_2e-1_connector-7.0_0.9_2e-1_ablation_20251018_062718.log
10
+ Timestamp: 2025-10-18 06:27:18
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.1_2e-1_connector-7.0_1.1_2e-1_ablation_20251018_062719.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.1_2e-1_connector-7.0_1.1_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.1_2e-1_connector-7.0_1.1_2e-1_ablation_20251018_062719.log
3
+ Timestamp: 2025-10-18 06:27:19
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.1_2e-1_connector-7.0_1.1_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.1_2e-1_connector-7.0_1.1_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.1_2e-1_connector-7.0_1.1_2e-1_ablation_20251018_062719.log
10
+ Timestamp: 2025-10-18 06:27:19
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.3_2e-1_connector-7.0_1.3_2e-1_ablation_20251018_062719.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.3_2e-1_connector-7.0_1.3_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.3_2e-1_connector-7.0_1.3_2e-1_ablation_20251018_062719.log
3
+ Timestamp: 2025-10-18 06:27:19
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.3_2e-1_connector-7.0_1.3_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.3_2e-1_connector-7.0_1.3_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.3_2e-1_connector-7.0_1.3_2e-1_ablation_20251018_062719.log
10
+ Timestamp: 2025-10-18 06:27:19
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.5_2e-1_connector-7.0_1.5_2e-1_ablation_20251018_062720.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.5_2e-1_connector-7.0_1.5_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.5_2e-1_connector-7.0_1.5_2e-1_ablation_20251018_062720.log
3
+ Timestamp: 2025-10-18 06:27:20
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.5_2e-1_connector-7.0_1.5_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.5_2e-1_connector-7.0_1.5_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.5_2e-1_connector-7.0_1.5_2e-1_ablation_20251018_062720.log
10
+ Timestamp: 2025-10-18 06:27:20
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.7_2e-1_connector-7.0_1.7_2e-1_ablation_20251018_062720.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.7_2e-1_connector-7.0_1.7_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.7_2e-1_connector-7.0_1.7_2e-1_ablation_20251018_062720.log
3
+ Timestamp: 2025-10-18 06:27:20
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.7_2e-1_connector-7.0_1.7_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.7_2e-1_connector-7.0_1.7_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.7_2e-1_connector-7.0_1.7_2e-1_ablation_20251018_062720.log
10
+ Timestamp: 2025-10-18 06:27:21
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.9_2e-1_connector-7.0_1.9_2e-1_ablation_20251018_062721.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.9_2e-1_connector-7.0_1.9_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.9_2e-1_connector-7.0_1.9_2e-1_ablation_20251018_062721.log
3
+ Timestamp: 2025-10-18 06:27:21
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.9_2e-1_connector-7.0_1.9_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.9_2e-1_connector-7.0_1.9_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_1.9_2e-1_connector-7.0_1.9_2e-1_ablation_20251018_062721.log
10
+ Timestamp: 2025-10-18 06:27:21
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.1_2e-1_connector-7.0_2.1_2e-1_ablation_20251018_062721.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.1_2e-1_connector-7.0_2.1_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.1_2e-1_connector-7.0_2.1_2e-1_ablation_20251018_062721.log
3
+ Timestamp: 2025-10-18 06:27:21
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.1_2e-1_connector-7.0_2.1_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.1_2e-1_connector-7.0_2.1_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.1_2e-1_connector-7.0_2.1_2e-1_ablation_20251018_062721.log
10
+ Timestamp: 2025-10-18 06:27:22
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.3_2e-1_connector-7.0_2.3_2e-1_ablation_20251018_062722.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.3_2e-1_connector-7.0_2.3_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.3_2e-1_connector-7.0_2.3_2e-1_ablation_20251018_062722.log
3
+ Timestamp: 2025-10-18 06:27:22
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.3_2e-1_connector-7.0_2.3_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.3_2e-1_connector-7.0_2.3_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.3_2e-1_connector-7.0_2.3_2e-1_ablation_20251018_062722.log
10
+ Timestamp: 2025-10-18 06:27:22
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.5_2e-1_connector-7.0_2.5_2e-1_ablation_20251018_062722.log ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.5_2e-1_connector-7.0_2.5_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.5_2e-1_connector-7.0_2.5_2e-1_ablation_20251018_062722.log
3
+ Timestamp: 2025-10-18 06:27:22
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.5_2e-1_connector-7.0_2.5_2e-1_ablation
6
+ python: can't open file '/s3-code/ywang29/ckpts/tinyllava/Oct17/scripts/apply_masks.py': [Errno 2] No such file or directory
7
+ bash: scripts/eval/mmmu.sh: No such file or directory
8
+ ==== EXPERIMENT COMPLETED: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.5_2e-1_connector-7.0_2.5_2e-1_ablation ====
9
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.5_2e-1_connector-7.0_2.5_2e-1_ablation_20251018_062722.log
10
+ Timestamp: 2025-10-18 06:27:23
11
+ =====================================
ckpts_oct17/eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.7_2e-1_connector-7.0_2.7_2e-1_ablation_20251018_062723.log ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ ==== STARTING EXPERIMENT: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.7_2e-1_connector-7.0_2.7_2e-1_ablation ====
2
+ Log File: eval_qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.7_2e-1_connector-7.0_2.7_2e-1_ablation_20251018_062723.log
3
+ Timestamp: 2025-10-18 06:27:23
4
+ =====================================
5
+ Processing: /s3-code/ywang29/ckpts/tinyllava/Oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-7.0_2.7_2e-1_connector-7.0_2.7_2e-1_ablation
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TinyLlavaForConditionalGeneration"
4
+ ],
5
+ "backward_type_connector": "normal",
6
+ "cache_dir": null,
7
+ "connector_type": "mlp2x_gelu",
8
+ "hidden_size": 896,
9
+ "ignore_index": -100,
10
+ "image_aspect_ratio": "square",
11
+ "image_token_index": -200,
12
+ "llm_model_name_or_path": "Qwen/Qwen2.5-0.5B",
13
+ "mask_model": [
14
+ "llm",
15
+ "connector"
16
+ ],
17
+ "mask_type_connector": "soft",
18
+ "model_type": "tinyllava",
19
+ "num_queries": 128,
20
+ "num_resampler_layers": 3,
21
+ "pad_token": "<|endoftext|>",
22
+ "pad_token_id": 151643,
23
+ "resampler_hidden_size": 768,
24
+ "sparsity_connector": null,
25
+ "subnet_type_connector": "global",
26
+ "temperature_connector": 1.7,
27
+ "text_config": {
28
+ "_name_or_path": "Qwen/Qwen2.5-0.5B",
29
+ "architectures": [
30
+ "Qwen2ForCausalLM"
31
+ ],
32
+ "backward_type": "normal",
33
+ "bos_token_id": 151643,
34
+ "eos_token_id": 151643,
35
+ "hidden_size": 896,
36
+ "intermediate_size": 4864,
37
+ "mask_type": "soft",
38
+ "masked_layers": "all",
39
+ "max_position_embeddings": 32768,
40
+ "max_window_layers": 24,
41
+ "model_type": "qwen2",
42
+ "num_attention_heads": 14,
43
+ "num_hidden_layers": 24,
44
+ "num_key_value_heads": 2,
45
+ "rope_theta": 1000000.0,
46
+ "sliding_window": 32768,
47
+ "subnet_mode": "both",
48
+ "subnet_type": "None",
49
+ "temperature_attn": 1.7,
50
+ "temperature_mlp": 1.7,
51
+ "tie_word_embeddings": true,
52
+ "torch_dtype": "bfloat16",
53
+ "use_mrope": false,
54
+ "use_sliding_window": false,
55
+ "vocab_size": 151936
56
+ },
57
+ "threshold_connector": null,
58
+ "tokenizer_model_max_length": 2048,
59
+ "tokenizer_name_or_path": "Qwen/Qwen2.5-0.5B",
60
+ "tokenizer_padding_side": "right",
61
+ "tokenizer_use_fast": false,
62
+ "torch_dtype": "bfloat16",
63
+ "transformers_version": "4.40.1",
64
+ "tune_type_connector": "full",
65
+ "tune_type_llm": "full",
66
+ "tune_type_vision_tower": "frozen",
67
+ "tune_vision_tower_from_layer": 0,
68
+ "use_cache": true,
69
+ "vision_config": {
70
+ "hidden_act": "gelu_pytorch_tanh",
71
+ "hidden_size": 1152,
72
+ "image_size": 384,
73
+ "intermediate_size": 4304,
74
+ "layer_norm_eps": 1e-06,
75
+ "model_name_or_path": "google/siglip-so400m-patch14-384",
76
+ "model_name_or_path2": "",
77
+ "model_type": "siglip_vision_model",
78
+ "num_attention_heads": 16,
79
+ "num_hidden_layers": 27,
80
+ "patch_size": 14
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "patch",
84
+ "vision_hidden_size": 1152,
85
+ "vision_model_name_or_path": "google/siglip-so400m-patch14-384",
86
+ "vision_model_name_or_path2": "",
87
+ "vocab_size": 151936
88
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151643,
4
+ "eos_token_id": 151643,
5
+ "transformers_version": "4.40.1",
6
+ "use_cache": false
7
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/log.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ 2025-10-17 08:18:26,213 | INFO: Total Parameters: 1283756736, Total Trainable Parameters: 359661568
2
+ 2025-10-17 08:18:26,221 | INFO: Trainable Parameters:
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:147a7497f2ebb45dc6c38bd1dedffa34b1938a61534f13828f3e661e38e3163d
3
+ size 2567640568
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/runs/Oct17_08-17-13_ywang29-vrdb-test1-worker-0/events.out.tfevents.1760689110.ywang29-vrdb-test1-worker-0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b88874beedc8bcab48786f21799d26cf14cc864c1dc1967883338e1955633ccc
3
+ size 116893
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/special_tokens_map.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ },
31
+ "unk_token": "<|endoftext|>"
32
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/tokenizer_config.json ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|endoftext|>",
201
+ "errors": "replace",
202
+ "model_max_length": 2048,
203
+ "pad_token": "<|endoftext|>",
204
+ "padding_side": "right",
205
+ "split_special_tokens": false,
206
+ "tokenizer_class": "Qwen2Tokenizer",
207
+ "unk_token": "<|endoftext|>"
208
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/trainer_state.json ADDED
@@ -0,0 +1,3670 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 520,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0019230769230769232,
13
+ "grad_norm": 0.43962132839848433,
14
+ "learning_rate": 0.0125,
15
+ "loss": 7.8159,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0038461538461538464,
20
+ "grad_norm": 0.4565264970706716,
21
+ "learning_rate": 0.025,
22
+ "loss": 7.0813,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.0057692307692307696,
27
+ "grad_norm": 0.26534792031522303,
28
+ "learning_rate": 0.037500000000000006,
29
+ "loss": 6.5681,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.007692307692307693,
34
+ "grad_norm": 0.15340078514641156,
35
+ "learning_rate": 0.05,
36
+ "loss": 4.9783,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.009615384615384616,
41
+ "grad_norm": 0.3382030787111079,
42
+ "learning_rate": 0.0625,
43
+ "loss": 4.5958,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.011538461538461539,
48
+ "grad_norm": 0.1299450415597302,
49
+ "learning_rate": 0.07500000000000001,
50
+ "loss": 4.9649,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.013461538461538462,
55
+ "grad_norm": 0.07749363911485645,
56
+ "learning_rate": 0.08750000000000001,
57
+ "loss": 3.4079,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.015384615384615385,
62
+ "grad_norm": 0.19758240155532028,
63
+ "learning_rate": 0.1,
64
+ "loss": 3.1923,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.01730769230769231,
69
+ "grad_norm": 0.03460659624842538,
70
+ "learning_rate": 0.1125,
71
+ "loss": 2.8155,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.019230769230769232,
76
+ "grad_norm": 0.029506617377956578,
77
+ "learning_rate": 0.125,
78
+ "loss": 2.3833,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.021153846153846155,
83
+ "grad_norm": 0.02111688152455683,
84
+ "learning_rate": 0.1375,
85
+ "loss": 2.357,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.023076923076923078,
90
+ "grad_norm": 0.027909163184843155,
91
+ "learning_rate": 0.15000000000000002,
92
+ "loss": 2.6349,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.025,
97
+ "grad_norm": 0.0167961992872781,
98
+ "learning_rate": 0.1625,
99
+ "loss": 2.1059,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.026923076923076925,
104
+ "grad_norm": 0.017144190080936636,
105
+ "learning_rate": 0.17500000000000002,
106
+ "loss": 2.1276,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.028846153846153848,
111
+ "grad_norm": 0.017149472634652465,
112
+ "learning_rate": 0.1875,
113
+ "loss": 2.3294,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.03076923076923077,
118
+ "grad_norm": 0.02290431799377215,
119
+ "learning_rate": 0.2,
120
+ "loss": 2.2412,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.032692307692307694,
125
+ "grad_norm": 0.0776205399250002,
126
+ "learning_rate": 0.1999980572931538,
127
+ "loss": 2.2649,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.03461538461538462,
132
+ "grad_norm": 0.08427472686802188,
133
+ "learning_rate": 0.19999222924809748,
134
+ "loss": 2.4645,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.03653846153846154,
139
+ "grad_norm": 0.2904409651880071,
140
+ "learning_rate": 0.19998251609127465,
141
+ "loss": 4.6016,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.038461538461538464,
146
+ "grad_norm": 0.07117107062913017,
147
+ "learning_rate": 0.19996891820008164,
148
+ "loss": 2.7929,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.04038461538461539,
153
+ "grad_norm": 0.05300705803334577,
154
+ "learning_rate": 0.19995143610285276,
155
+ "loss": 3.2124,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.04230769230769231,
160
+ "grad_norm": 0.039699136066479956,
161
+ "learning_rate": 0.19993007047883987,
162
+ "loss": 2.5256,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.04423076923076923,
167
+ "grad_norm": 0.017606880929474766,
168
+ "learning_rate": 0.1999048221581858,
169
+ "loss": 2.2677,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.046153846153846156,
174
+ "grad_norm": 0.026558842416946657,
175
+ "learning_rate": 0.19987569212189224,
176
+ "loss": 2.6303,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.04807692307692308,
181
+ "grad_norm": 0.02563381268462812,
182
+ "learning_rate": 0.19984268150178167,
183
+ "loss": 2.3081,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.05,
188
+ "grad_norm": 0.013209202488013532,
189
+ "learning_rate": 0.1998057915804532,
190
+ "loss": 2.2071,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.051923076923076926,
195
+ "grad_norm": 0.017399859578397278,
196
+ "learning_rate": 0.1997650237912329,
197
+ "loss": 2.0107,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.05384615384615385,
202
+ "grad_norm": 0.010676025029797469,
203
+ "learning_rate": 0.199720379718118,
204
+ "loss": 1.9714,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.05576923076923077,
209
+ "grad_norm": 0.015350226061628867,
210
+ "learning_rate": 0.19967186109571552,
211
+ "loss": 1.9705,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.057692307692307696,
216
+ "grad_norm": 0.01671849980026423,
217
+ "learning_rate": 0.19961946980917455,
218
+ "loss": 2.5243,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.05961538461538462,
223
+ "grad_norm": 0.009793921347140364,
224
+ "learning_rate": 0.1995632078941134,
225
+ "loss": 1.9463,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.06153846153846154,
230
+ "grad_norm": 0.04250056185239384,
231
+ "learning_rate": 0.19950307753654017,
232
+ "loss": 2.781,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.06346153846153846,
237
+ "grad_norm": 0.02342504110070967,
238
+ "learning_rate": 0.19943908107276798,
239
+ "loss": 1.9901,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.06538461538461539,
244
+ "grad_norm": 0.008516983738821215,
245
+ "learning_rate": 0.19937122098932428,
246
+ "loss": 1.9084,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.0673076923076923,
251
+ "grad_norm": 0.013513648575825075,
252
+ "learning_rate": 0.19929949992285395,
253
+ "loss": 1.9376,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.06923076923076923,
258
+ "grad_norm": 0.014137521727245587,
259
+ "learning_rate": 0.19922392066001723,
260
+ "loss": 2.0509,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.07115384615384615,
265
+ "grad_norm": 0.02335657031196121,
266
+ "learning_rate": 0.19914448613738106,
267
+ "loss": 2.3764,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.07307692307692308,
272
+ "grad_norm": 0.020772001259702376,
273
+ "learning_rate": 0.1990611994413053,
274
+ "loss": 2.0886,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.075,
279
+ "grad_norm": 0.007049709090839739,
280
+ "learning_rate": 0.19897406380782262,
281
+ "loss": 1.9498,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.07692307692307693,
286
+ "grad_norm": 0.016449487690420556,
287
+ "learning_rate": 0.19888308262251286,
288
+ "loss": 1.8843,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.07884615384615384,
293
+ "grad_norm": 0.00966195934277566,
294
+ "learning_rate": 0.19878825942037148,
295
+ "loss": 1.8508,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.08076923076923077,
300
+ "grad_norm": 0.009838489709262746,
301
+ "learning_rate": 0.19868959788567211,
302
+ "loss": 1.9913,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.08269230769230769,
307
+ "grad_norm": 0.015211827148146388,
308
+ "learning_rate": 0.1985871018518236,
309
+ "loss": 2.0994,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.08461538461538462,
314
+ "grad_norm": 0.008877769806496752,
315
+ "learning_rate": 0.19848077530122082,
316
+ "loss": 2.1957,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.08653846153846154,
321
+ "grad_norm": 0.007848779222195778,
322
+ "learning_rate": 0.19837062236509015,
323
+ "loss": 1.826,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.08846153846153847,
328
+ "grad_norm": 0.013017740266987338,
329
+ "learning_rate": 0.19825664732332884,
330
+ "loss": 2.2393,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.09038461538461538,
335
+ "grad_norm": 0.008240739311325988,
336
+ "learning_rate": 0.19813885460433878,
337
+ "loss": 1.8304,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.09230769230769231,
342
+ "grad_norm": 0.007317875200128298,
343
+ "learning_rate": 0.19801724878485438,
344
+ "loss": 1.8094,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.09423076923076923,
349
+ "grad_norm": 0.008483845155024662,
350
+ "learning_rate": 0.19789183458976486,
351
+ "loss": 1.7777,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.09615384615384616,
356
+ "grad_norm": 0.007089011209416157,
357
+ "learning_rate": 0.19776261689193048,
358
+ "loss": 1.7562,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.09807692307692308,
363
+ "grad_norm": 0.006034570951852243,
364
+ "learning_rate": 0.19762960071199334,
365
+ "loss": 1.679,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.1,
370
+ "grad_norm": 0.006985484724489631,
371
+ "learning_rate": 0.19749279121818236,
372
+ "loss": 1.8468,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.10192307692307692,
377
+ "grad_norm": 0.0077857638590927165,
378
+ "learning_rate": 0.19735219372611235,
379
+ "loss": 1.8219,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.10384615384615385,
384
+ "grad_norm": 0.005424431466869681,
385
+ "learning_rate": 0.19720781369857746,
386
+ "loss": 1.6618,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.10576923076923077,
391
+ "grad_norm": 0.006775142447487382,
392
+ "learning_rate": 0.1970596567453391,
393
+ "loss": 1.6498,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.1076923076923077,
398
+ "grad_norm": 0.006855723481981793,
399
+ "learning_rate": 0.1969077286229078,
400
+ "loss": 1.8158,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.10961538461538461,
405
+ "grad_norm": 0.006728639141520153,
406
+ "learning_rate": 0.19675203523431964,
407
+ "loss": 1.6403,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.11153846153846154,
412
+ "grad_norm": 0.005144455211432859,
413
+ "learning_rate": 0.19659258262890683,
414
+ "loss": 1.7833,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.11346153846153846,
419
+ "grad_norm": 0.008371659863437526,
420
+ "learning_rate": 0.19642937700206278,
421
+ "loss": 1.8395,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.11538461538461539,
426
+ "grad_norm": 0.007172251875338679,
427
+ "learning_rate": 0.19626242469500121,
428
+ "loss": 1.7146,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.11730769230769231,
433
+ "grad_norm": 0.007422352782010027,
434
+ "learning_rate": 0.19609173219450998,
435
+ "loss": 1.9583,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.11923076923076924,
440
+ "grad_norm": 0.006557810736509136,
441
+ "learning_rate": 0.19591730613269878,
442
+ "loss": 1.6781,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.12115384615384615,
447
+ "grad_norm": 0.005694810873534034,
448
+ "learning_rate": 0.19573915328674182,
449
+ "loss": 1.6745,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.12307692307692308,
454
+ "grad_norm": 0.005886951265824137,
455
+ "learning_rate": 0.1955572805786141,
456
+ "loss": 1.6863,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.125,
461
+ "grad_norm": 0.006522252515055469,
462
+ "learning_rate": 0.1953716950748227,
463
+ "loss": 1.7182,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.12692307692307692,
468
+ "grad_norm": 0.006476783772659108,
469
+ "learning_rate": 0.19518240398613226,
470
+ "loss": 1.6599,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.12884615384615383,
475
+ "grad_norm": 0.006666730512418093,
476
+ "learning_rate": 0.1949894146672846,
477
+ "loss": 1.5027,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.13076923076923078,
482
+ "grad_norm": 0.005156415743585036,
483
+ "learning_rate": 0.1947927346167132,
484
+ "loss": 1.5553,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.1326923076923077,
489
+ "grad_norm": 0.00925563237418746,
490
+ "learning_rate": 0.1945923714762516,
491
+ "loss": 1.5557,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.1346153846153846,
496
+ "grad_norm": 0.0066292027026927415,
497
+ "learning_rate": 0.19438833303083677,
498
+ "loss": 1.6139,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.13653846153846153,
503
+ "grad_norm": 0.005377919049465734,
504
+ "learning_rate": 0.19418062720820636,
505
+ "loss": 1.5069,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.13846153846153847,
510
+ "grad_norm": 0.006722866957177672,
511
+ "learning_rate": 0.19396926207859086,
512
+ "loss": 1.6698,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.14038461538461539,
517
+ "grad_norm": 0.005657512829681465,
518
+ "learning_rate": 0.19375424585439993,
519
+ "loss": 1.4915,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.1423076923076923,
524
+ "grad_norm": 0.006357916421070106,
525
+ "learning_rate": 0.1935355868899034,
526
+ "loss": 1.6298,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.14423076923076922,
531
+ "grad_norm": 0.005356680536328098,
532
+ "learning_rate": 0.19331329368090666,
533
+ "loss": 1.5035,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.14615384615384616,
538
+ "grad_norm": 0.008006470059962833,
539
+ "learning_rate": 0.19308737486442043,
540
+ "loss": 1.9297,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.14807692307692308,
545
+ "grad_norm": 0.0071590363492128346,
546
+ "learning_rate": 0.19285783921832536,
547
+ "loss": 1.4254,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.15,
552
+ "grad_norm": 0.004849803746589159,
553
+ "learning_rate": 0.19262469566103088,
554
+ "loss": 1.552,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.1519230769230769,
559
+ "grad_norm": 0.005024691697437532,
560
+ "learning_rate": 0.19238795325112867,
561
+ "loss": 1.5345,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.15384615384615385,
566
+ "grad_norm": 0.010829571319044851,
567
+ "learning_rate": 0.19214762118704076,
568
+ "loss": 1.9422,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.15576923076923077,
573
+ "grad_norm": 0.006644890147771466,
574
+ "learning_rate": 0.19190370880666208,
575
+ "loss": 1.7066,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.1576923076923077,
580
+ "grad_norm": 0.005404604994645463,
581
+ "learning_rate": 0.19165622558699763,
582
+ "loss": 1.6016,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.1596153846153846,
587
+ "grad_norm": 0.005983372135820056,
588
+ "learning_rate": 0.19140518114379435,
589
+ "loss": 1.6628,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.16153846153846155,
594
+ "grad_norm": 0.005720595354668528,
595
+ "learning_rate": 0.19115058523116735,
596
+ "loss": 1.6398,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.16346153846153846,
601
+ "grad_norm": 0.006711848246282018,
602
+ "learning_rate": 0.1908924477412211,
603
+ "loss": 1.6371,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.16538461538461538,
608
+ "grad_norm": 0.0074539981815448415,
609
+ "learning_rate": 0.19063077870366502,
610
+ "loss": 1.6884,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.1673076923076923,
615
+ "grad_norm": 0.00989742251671178,
616
+ "learning_rate": 0.1903655882854237,
617
+ "loss": 1.9109,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.16923076923076924,
622
+ "grad_norm": 0.013020192526349603,
623
+ "learning_rate": 0.19009688679024192,
624
+ "loss": 1.9338,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.17115384615384616,
629
+ "grad_norm": 0.006015920749467016,
630
+ "learning_rate": 0.18982468465828442,
631
+ "loss": 1.5941,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.17307692307692307,
636
+ "grad_norm": 0.007068287630888444,
637
+ "learning_rate": 0.1895489924657301,
638
+ "loss": 1.5302,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.175,
643
+ "grad_norm": 0.005139167085278699,
644
+ "learning_rate": 0.18926982092436118,
645
+ "loss": 1.6049,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.17692307692307693,
650
+ "grad_norm": 0.007043781866498074,
651
+ "learning_rate": 0.18898718088114688,
652
+ "loss": 1.5375,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.17884615384615385,
657
+ "grad_norm": 0.005745829486193244,
658
+ "learning_rate": 0.18870108331782218,
659
+ "loss": 1.5431,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.18076923076923077,
664
+ "grad_norm": 0.00665239964726687,
665
+ "learning_rate": 0.18841153935046098,
666
+ "loss": 1.6665,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.18269230769230768,
671
+ "grad_norm": 0.005470238738386594,
672
+ "learning_rate": 0.18811856022904425,
673
+ "loss": 1.5228,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.18461538461538463,
678
+ "grad_norm": 0.005342487067427777,
679
+ "learning_rate": 0.18782215733702287,
680
+ "loss": 1.5187,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.18653846153846154,
685
+ "grad_norm": 0.006758235711068036,
686
+ "learning_rate": 0.18752234219087538,
687
+ "loss": 1.4925,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.18846153846153846,
692
+ "grad_norm": 0.004554773248951545,
693
+ "learning_rate": 0.18721912643966054,
694
+ "loss": 1.4882,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.19038461538461537,
699
+ "grad_norm": 0.004992541136987346,
700
+ "learning_rate": 0.18691252186456464,
701
+ "loss": 1.5067,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.19230769230769232,
706
+ "grad_norm": 0.00873324790845758,
707
+ "learning_rate": 0.1866025403784439,
708
+ "loss": 1.7537,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.19423076923076923,
713
+ "grad_norm": 0.0052358066156105365,
714
+ "learning_rate": 0.18628919402536132,
715
+ "loss": 1.4993,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.19615384615384615,
720
+ "grad_norm": 0.005434214414672433,
721
+ "learning_rate": 0.18597249498011903,
722
+ "loss": 1.5126,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.19807692307692307,
727
+ "grad_norm": 0.004599633671126387,
728
+ "learning_rate": 0.18565245554778517,
729
+ "loss": 1.4414,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.2,
734
+ "grad_norm": 0.005377560701431235,
735
+ "learning_rate": 0.18532908816321558,
736
+ "loss": 1.5253,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.20192307692307693,
741
+ "grad_norm": 0.004868292325841549,
742
+ "learning_rate": 0.18500240539057092,
743
+ "loss": 1.5209,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.20384615384615384,
748
+ "grad_norm": 0.008280881226815754,
749
+ "learning_rate": 0.18467241992282843,
750
+ "loss": 1.6906,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.20576923076923076,
755
+ "grad_norm": 0.005970502210517993,
756
+ "learning_rate": 0.18433914458128858,
757
+ "loss": 1.666,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.2076923076923077,
762
+ "grad_norm": 0.006090082015069985,
763
+ "learning_rate": 0.18400259231507718,
764
+ "loss": 1.468,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.20961538461538462,
769
+ "grad_norm": 0.007446535104954824,
770
+ "learning_rate": 0.18366277620064198,
771
+ "loss": 1.6369,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.21153846153846154,
776
+ "grad_norm": 0.006504693217160285,
777
+ "learning_rate": 0.1833197094412449,
778
+ "loss": 1.6674,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.21346153846153845,
783
+ "grad_norm": 0.00602349680918933,
784
+ "learning_rate": 0.18297340536644877,
785
+ "loss": 1.6725,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.2153846153846154,
790
+ "grad_norm": 0.005233237619822541,
791
+ "learning_rate": 0.1826238774315995,
792
+ "loss": 1.5427,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.2173076923076923,
797
+ "grad_norm": 0.004637337585087387,
798
+ "learning_rate": 0.18227113921730334,
799
+ "loss": 1.396,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.21923076923076923,
804
+ "grad_norm": 0.005530922509985472,
805
+ "learning_rate": 0.1819152044288992,
806
+ "loss": 1.5096,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.22115384615384615,
811
+ "grad_norm": 0.0052301194615377794,
812
+ "learning_rate": 0.18155608689592603,
813
+ "loss": 1.6521,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.2230769230769231,
818
+ "grad_norm": 0.004693128126444144,
819
+ "learning_rate": 0.18119380057158568,
820
+ "loss": 1.6158,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.225,
825
+ "grad_norm": 0.005306705929593196,
826
+ "learning_rate": 0.18082835953220056,
827
+ "loss": 1.6137,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.22692307692307692,
832
+ "grad_norm": 0.0044518594670451325,
833
+ "learning_rate": 0.18045977797666685,
834
+ "loss": 1.484,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.22884615384615384,
839
+ "grad_norm": 0.004734720873448768,
840
+ "learning_rate": 0.1800880702259028,
841
+ "loss": 1.4231,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.23076923076923078,
846
+ "grad_norm": 0.005266254187638369,
847
+ "learning_rate": 0.17971325072229227,
848
+ "loss": 1.4494,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.2326923076923077,
853
+ "grad_norm": 0.006383736724371343,
854
+ "learning_rate": 0.17933533402912352,
855
+ "loss": 1.5214,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.23461538461538461,
860
+ "grad_norm": 0.00492055255177075,
861
+ "learning_rate": 0.17895433483002354,
862
+ "loss": 1.3969,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.23653846153846153,
867
+ "grad_norm": 0.006716463406044716,
868
+ "learning_rate": 0.17857026792838737,
869
+ "loss": 1.7138,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 0.23846153846153847,
874
+ "grad_norm": 0.006410590043198477,
875
+ "learning_rate": 0.178183148246803,
876
+ "loss": 1.5079,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 0.2403846153846154,
881
+ "grad_norm": 0.005425927822436429,
882
+ "learning_rate": 0.1777929908264715,
883
+ "loss": 1.4701,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 0.2423076923076923,
888
+ "grad_norm": 0.0051191319289095455,
889
+ "learning_rate": 0.17739981082662276,
890
+ "loss": 1.5937,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 0.24423076923076922,
895
+ "grad_norm": 0.006098352898744862,
896
+ "learning_rate": 0.1770036235239263,
897
+ "loss": 1.4542,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 0.24615384615384617,
902
+ "grad_norm": 0.005209028163915202,
903
+ "learning_rate": 0.1766044443118978,
904
+ "loss": 1.5096,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 0.24807692307692308,
909
+ "grad_norm": 0.003964791585808975,
910
+ "learning_rate": 0.17620228870030108,
911
+ "loss": 1.4208,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 0.25,
916
+ "grad_norm": 0.004819018194236482,
917
+ "learning_rate": 0.1757971723145453,
918
+ "loss": 1.4862,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 0.2519230769230769,
923
+ "grad_norm": 0.006206123315304021,
924
+ "learning_rate": 0.175389110895078,
925
+ "loss": 1.5631,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 0.25384615384615383,
930
+ "grad_norm": 0.005149376271850292,
931
+ "learning_rate": 0.17497812029677343,
932
+ "loss": 1.5374,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 0.25576923076923075,
937
+ "grad_norm": 0.005200608392851684,
938
+ "learning_rate": 0.17456421648831655,
939
+ "loss": 1.4404,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 0.25769230769230766,
944
+ "grad_norm": 0.004738669891883301,
945
+ "learning_rate": 0.17414741555158267,
946
+ "loss": 1.5196,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 0.25961538461538464,
951
+ "grad_norm": 0.005148904189984861,
952
+ "learning_rate": 0.1737277336810124,
953
+ "loss": 1.6007,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 0.26153846153846155,
958
+ "grad_norm": 0.004762039287477681,
959
+ "learning_rate": 0.17330518718298263,
960
+ "loss": 1.5168,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 0.26346153846153847,
965
+ "grad_norm": 0.006294002973782846,
966
+ "learning_rate": 0.17287979247517285,
967
+ "loss": 1.4299,
968
+ "step": 137
969
+ },
970
+ {
971
+ "epoch": 0.2653846153846154,
972
+ "grad_norm": 0.005049294732324241,
973
+ "learning_rate": 0.17245156608592727,
974
+ "loss": 1.4334,
975
+ "step": 138
976
+ },
977
+ {
978
+ "epoch": 0.2673076923076923,
979
+ "grad_norm": 0.004896138904157698,
980
+ "learning_rate": 0.17202052465361267,
981
+ "loss": 1.4495,
982
+ "step": 139
983
+ },
984
+ {
985
+ "epoch": 0.2692307692307692,
986
+ "grad_norm": 0.005469191385153808,
987
+ "learning_rate": 0.17158668492597184,
988
+ "loss": 1.6109,
989
+ "step": 140
990
+ },
991
+ {
992
+ "epoch": 0.27115384615384613,
993
+ "grad_norm": 0.005883849042069001,
994
+ "learning_rate": 0.17115006375947303,
995
+ "loss": 1.5651,
996
+ "step": 141
997
+ },
998
+ {
999
+ "epoch": 0.27307692307692305,
1000
+ "grad_norm": 0.006445635177801752,
1001
+ "learning_rate": 0.17071067811865476,
1002
+ "loss": 1.6486,
1003
+ "step": 142
1004
+ },
1005
+ {
1006
+ "epoch": 0.275,
1007
+ "grad_norm": 0.006737590764970197,
1008
+ "learning_rate": 0.17026854507546693,
1009
+ "loss": 1.4809,
1010
+ "step": 143
1011
+ },
1012
+ {
1013
+ "epoch": 0.27692307692307694,
1014
+ "grad_norm": 0.006136643339042509,
1015
+ "learning_rate": 0.1698236818086073,
1016
+ "loss": 1.4257,
1017
+ "step": 144
1018
+ },
1019
+ {
1020
+ "epoch": 0.27884615384615385,
1021
+ "grad_norm": 0.004624453843412489,
1022
+ "learning_rate": 0.16937610560285418,
1023
+ "loss": 1.3513,
1024
+ "step": 145
1025
+ },
1026
+ {
1027
+ "epoch": 0.28076923076923077,
1028
+ "grad_norm": 0.005646188796662644,
1029
+ "learning_rate": 0.1689258338483947,
1030
+ "loss": 1.6828,
1031
+ "step": 146
1032
+ },
1033
+ {
1034
+ "epoch": 0.2826923076923077,
1035
+ "grad_norm": 0.005303245549838439,
1036
+ "learning_rate": 0.16847288404014937,
1037
+ "loss": 1.3954,
1038
+ "step": 147
1039
+ },
1040
+ {
1041
+ "epoch": 0.2846153846153846,
1042
+ "grad_norm": 0.004529711918179032,
1043
+ "learning_rate": 0.16801727377709194,
1044
+ "loss": 1.4387,
1045
+ "step": 148
1046
+ },
1047
+ {
1048
+ "epoch": 0.2865384615384615,
1049
+ "grad_norm": 0.005546577970652367,
1050
+ "learning_rate": 0.16755902076156604,
1051
+ "loss": 1.3877,
1052
+ "step": 149
1053
+ },
1054
+ {
1055
+ "epoch": 0.28846153846153844,
1056
+ "grad_norm": 0.006910112853420524,
1057
+ "learning_rate": 0.16709814279859703,
1058
+ "loss": 1.6198,
1059
+ "step": 150
1060
+ },
1061
+ {
1062
+ "epoch": 0.2903846153846154,
1063
+ "grad_norm": 0.0043634320178486535,
1064
+ "learning_rate": 0.1666346577952004,
1065
+ "loss": 1.412,
1066
+ "step": 151
1067
+ },
1068
+ {
1069
+ "epoch": 0.2923076923076923,
1070
+ "grad_norm": 0.006489334489083526,
1071
+ "learning_rate": 0.16616858375968596,
1072
+ "loss": 1.3784,
1073
+ "step": 152
1074
+ },
1075
+ {
1076
+ "epoch": 0.29423076923076924,
1077
+ "grad_norm": 0.006362415101858187,
1078
+ "learning_rate": 0.16569993880095807,
1079
+ "loss": 1.4345,
1080
+ "step": 153
1081
+ },
1082
+ {
1083
+ "epoch": 0.29615384615384616,
1084
+ "grad_norm": 0.00429250739710453,
1085
+ "learning_rate": 0.16522874112781213,
1086
+ "loss": 1.511,
1087
+ "step": 154
1088
+ },
1089
+ {
1090
+ "epoch": 0.2980769230769231,
1091
+ "grad_norm": 0.005746047748370822,
1092
+ "learning_rate": 0.16475500904822704,
1093
+ "loss": 1.4225,
1094
+ "step": 155
1095
+ },
1096
+ {
1097
+ "epoch": 0.3,
1098
+ "grad_norm": 0.005844450834838985,
1099
+ "learning_rate": 0.16427876096865393,
1100
+ "loss": 1.4509,
1101
+ "step": 156
1102
+ },
1103
+ {
1104
+ "epoch": 0.3019230769230769,
1105
+ "grad_norm": 0.004975343698957949,
1106
+ "learning_rate": 0.16380001539330089,
1107
+ "loss": 1.7,
1108
+ "step": 157
1109
+ },
1110
+ {
1111
+ "epoch": 0.3038461538461538,
1112
+ "grad_norm": 0.006991355067128036,
1113
+ "learning_rate": 0.163318790923414,
1114
+ "loss": 1.4283,
1115
+ "step": 158
1116
+ },
1117
+ {
1118
+ "epoch": 0.3057692307692308,
1119
+ "grad_norm": 0.004963334499761017,
1120
+ "learning_rate": 0.16283510625655473,
1121
+ "loss": 1.4472,
1122
+ "step": 159
1123
+ },
1124
+ {
1125
+ "epoch": 0.3076923076923077,
1126
+ "grad_norm": 0.0047525778394396714,
1127
+ "learning_rate": 0.16234898018587338,
1128
+ "loss": 1.4895,
1129
+ "step": 160
1130
+ },
1131
+ {
1132
+ "epoch": 0.3096153846153846,
1133
+ "grad_norm": 0.005385963045094978,
1134
+ "learning_rate": 0.16186043159937882,
1135
+ "loss": 1.475,
1136
+ "step": 161
1137
+ },
1138
+ {
1139
+ "epoch": 0.31153846153846154,
1140
+ "grad_norm": 0.005794363331681811,
1141
+ "learning_rate": 0.16136947947920477,
1142
+ "loss": 1.6036,
1143
+ "step": 162
1144
+ },
1145
+ {
1146
+ "epoch": 0.31346153846153846,
1147
+ "grad_norm": 0.005739269683633919,
1148
+ "learning_rate": 0.16087614290087207,
1149
+ "loss": 1.3393,
1150
+ "step": 163
1151
+ },
1152
+ {
1153
+ "epoch": 0.3153846153846154,
1154
+ "grad_norm": 0.005097443247732428,
1155
+ "learning_rate": 0.16038044103254775,
1156
+ "loss": 1.3022,
1157
+ "step": 164
1158
+ },
1159
+ {
1160
+ "epoch": 0.3173076923076923,
1161
+ "grad_norm": 0.004990016261688464,
1162
+ "learning_rate": 0.15988239313430005,
1163
+ "loss": 1.4421,
1164
+ "step": 165
1165
+ },
1166
+ {
1167
+ "epoch": 0.3192307692307692,
1168
+ "grad_norm": 0.005132623079008562,
1169
+ "learning_rate": 0.15938201855735015,
1170
+ "loss": 1.4408,
1171
+ "step": 166
1172
+ },
1173
+ {
1174
+ "epoch": 0.3211538461538462,
1175
+ "grad_norm": 0.005996703719727143,
1176
+ "learning_rate": 0.15887933674332047,
1177
+ "loss": 1.4371,
1178
+ "step": 167
1179
+ },
1180
+ {
1181
+ "epoch": 0.3230769230769231,
1182
+ "grad_norm": 0.004834974813698066,
1183
+ "learning_rate": 0.158374367223479,
1184
+ "loss": 1.3684,
1185
+ "step": 168
1186
+ },
1187
+ {
1188
+ "epoch": 0.325,
1189
+ "grad_norm": 0.004193615747650312,
1190
+ "learning_rate": 0.1578671296179806,
1191
+ "loss": 1.4436,
1192
+ "step": 169
1193
+ },
1194
+ {
1195
+ "epoch": 0.3269230769230769,
1196
+ "grad_norm": 0.006279609199566032,
1197
+ "learning_rate": 0.15735764363510463,
1198
+ "loss": 1.5109,
1199
+ "step": 170
1200
+ },
1201
+ {
1202
+ "epoch": 0.32884615384615384,
1203
+ "grad_norm": 0.005602664490223233,
1204
+ "learning_rate": 0.15684592907048925,
1205
+ "loss": 1.3748,
1206
+ "step": 171
1207
+ },
1208
+ {
1209
+ "epoch": 0.33076923076923076,
1210
+ "grad_norm": 0.004488241362940413,
1211
+ "learning_rate": 0.1563320058063622,
1212
+ "loss": 1.4417,
1213
+ "step": 172
1214
+ },
1215
+ {
1216
+ "epoch": 0.3326923076923077,
1217
+ "grad_norm": 0.0047776634460675385,
1218
+ "learning_rate": 0.15581589381076844,
1219
+ "loss": 1.368,
1220
+ "step": 173
1221
+ },
1222
+ {
1223
+ "epoch": 0.3346153846153846,
1224
+ "grad_norm": 0.005829143545890202,
1225
+ "learning_rate": 0.15529761313679394,
1226
+ "loss": 1.4592,
1227
+ "step": 174
1228
+ },
1229
+ {
1230
+ "epoch": 0.33653846153846156,
1231
+ "grad_norm": 0.004858986501599738,
1232
+ "learning_rate": 0.15477718392178716,
1233
+ "loss": 1.3522,
1234
+ "step": 175
1235
+ },
1236
+ {
1237
+ "epoch": 0.3384615384615385,
1238
+ "grad_norm": 0.005886157820841517,
1239
+ "learning_rate": 0.15425462638657594,
1240
+ "loss": 1.6066,
1241
+ "step": 176
1242
+ },
1243
+ {
1244
+ "epoch": 0.3403846153846154,
1245
+ "grad_norm": 0.00589056998996616,
1246
+ "learning_rate": 0.1537299608346824,
1247
+ "loss": 1.4628,
1248
+ "step": 177
1249
+ },
1250
+ {
1251
+ "epoch": 0.3423076923076923,
1252
+ "grad_norm": 0.004860225049341871,
1253
+ "learning_rate": 0.15320320765153367,
1254
+ "loss": 1.4317,
1255
+ "step": 178
1256
+ },
1257
+ {
1258
+ "epoch": 0.34423076923076923,
1259
+ "grad_norm": 0.004439546416827562,
1260
+ "learning_rate": 0.15267438730367008,
1261
+ "loss": 1.5128,
1262
+ "step": 179
1263
+ },
1264
+ {
1265
+ "epoch": 0.34615384615384615,
1266
+ "grad_norm": 0.0059684433101612555,
1267
+ "learning_rate": 0.1521435203379498,
1268
+ "loss": 1.4201,
1269
+ "step": 180
1270
+ },
1271
+ {
1272
+ "epoch": 0.34807692307692306,
1273
+ "grad_norm": 0.004340660435216258,
1274
+ "learning_rate": 0.15161062738075068,
1275
+ "loss": 1.3917,
1276
+ "step": 181
1277
+ },
1278
+ {
1279
+ "epoch": 0.35,
1280
+ "grad_norm": 0.004257557860940274,
1281
+ "learning_rate": 0.1510757291371686,
1282
+ "loss": 1.4069,
1283
+ "step": 182
1284
+ },
1285
+ {
1286
+ "epoch": 0.35192307692307695,
1287
+ "grad_norm": 0.005639353585625431,
1288
+ "learning_rate": 0.1505388463902131,
1289
+ "loss": 1.4372,
1290
+ "step": 183
1291
+ },
1292
+ {
1293
+ "epoch": 0.35384615384615387,
1294
+ "grad_norm": 0.004396333522794009,
1295
+ "learning_rate": 0.15000000000000002,
1296
+ "loss": 1.3465,
1297
+ "step": 184
1298
+ },
1299
+ {
1300
+ "epoch": 0.3557692307692308,
1301
+ "grad_norm": 0.004392134116413106,
1302
+ "learning_rate": 0.14945921090294076,
1303
+ "loss": 1.5278,
1304
+ "step": 185
1305
+ },
1306
+ {
1307
+ "epoch": 0.3576923076923077,
1308
+ "grad_norm": 0.004352645728204937,
1309
+ "learning_rate": 0.14891650011092894,
1310
+ "loss": 1.3571,
1311
+ "step": 186
1312
+ },
1313
+ {
1314
+ "epoch": 0.3596153846153846,
1315
+ "grad_norm": 0.0051957385329964995,
1316
+ "learning_rate": 0.14837188871052398,
1317
+ "loss": 1.3775,
1318
+ "step": 187
1319
+ },
1320
+ {
1321
+ "epoch": 0.36153846153846153,
1322
+ "grad_norm": 0.004642775148278487,
1323
+ "learning_rate": 0.14782539786213184,
1324
+ "loss": 1.4542,
1325
+ "step": 188
1326
+ },
1327
+ {
1328
+ "epoch": 0.36346153846153845,
1329
+ "grad_norm": 0.0042437486115803145,
1330
+ "learning_rate": 0.1472770487991827,
1331
+ "loss": 1.4671,
1332
+ "step": 189
1333
+ },
1334
+ {
1335
+ "epoch": 0.36538461538461536,
1336
+ "grad_norm": 0.004695628817689261,
1337
+ "learning_rate": 0.1467268628273062,
1338
+ "loss": 1.3776,
1339
+ "step": 190
1340
+ },
1341
+ {
1342
+ "epoch": 0.36730769230769234,
1343
+ "grad_norm": 0.0041549444745122175,
1344
+ "learning_rate": 0.1461748613235034,
1345
+ "loss": 1.3301,
1346
+ "step": 191
1347
+ },
1348
+ {
1349
+ "epoch": 0.36923076923076925,
1350
+ "grad_norm": 0.004654262785627284,
1351
+ "learning_rate": 0.1456210657353163,
1352
+ "loss": 1.4296,
1353
+ "step": 192
1354
+ },
1355
+ {
1356
+ "epoch": 0.37115384615384617,
1357
+ "grad_norm": 0.005293812452344632,
1358
+ "learning_rate": 0.14506549757999454,
1359
+ "loss": 1.5441,
1360
+ "step": 193
1361
+ },
1362
+ {
1363
+ "epoch": 0.3730769230769231,
1364
+ "grad_norm": 0.007179093940230065,
1365
+ "learning_rate": 0.14450817844365924,
1366
+ "loss": 1.3926,
1367
+ "step": 194
1368
+ },
1369
+ {
1370
+ "epoch": 0.375,
1371
+ "grad_norm": 0.00427825241149333,
1372
+ "learning_rate": 0.1439491299804645,
1373
+ "loss": 1.446,
1374
+ "step": 195
1375
+ },
1376
+ {
1377
+ "epoch": 0.3769230769230769,
1378
+ "grad_norm": 0.005138828447024368,
1379
+ "learning_rate": 0.14338837391175582,
1380
+ "loss": 1.4033,
1381
+ "step": 196
1382
+ },
1383
+ {
1384
+ "epoch": 0.37884615384615383,
1385
+ "grad_norm": 0.004432347846415011,
1386
+ "learning_rate": 0.14282593202522628,
1387
+ "loss": 1.372,
1388
+ "step": 197
1389
+ },
1390
+ {
1391
+ "epoch": 0.38076923076923075,
1392
+ "grad_norm": 0.004720740560216542,
1393
+ "learning_rate": 0.14226182617406996,
1394
+ "loss": 1.4486,
1395
+ "step": 198
1396
+ },
1397
+ {
1398
+ "epoch": 0.38269230769230766,
1399
+ "grad_norm": 0.004293098593386714,
1400
+ "learning_rate": 0.14169607827613281,
1401
+ "loss": 1.3655,
1402
+ "step": 199
1403
+ },
1404
+ {
1405
+ "epoch": 0.38461538461538464,
1406
+ "grad_norm": 0.005097745891661831,
1407
+ "learning_rate": 0.14112871031306118,
1408
+ "loss": 1.462,
1409
+ "step": 200
1410
+ },
1411
+ {
1412
+ "epoch": 0.38653846153846155,
1413
+ "grad_norm": 0.004278939462254616,
1414
+ "learning_rate": 0.1405597443294475,
1415
+ "loss": 1.4772,
1416
+ "step": 201
1417
+ },
1418
+ {
1419
+ "epoch": 0.38846153846153847,
1420
+ "grad_norm": 0.00450293598936501,
1421
+ "learning_rate": 0.13998920243197407,
1422
+ "loss": 1.3583,
1423
+ "step": 202
1424
+ },
1425
+ {
1426
+ "epoch": 0.3903846153846154,
1427
+ "grad_norm": 0.00452036977149978,
1428
+ "learning_rate": 0.13941710678855396,
1429
+ "loss": 1.4044,
1430
+ "step": 203
1431
+ },
1432
+ {
1433
+ "epoch": 0.3923076923076923,
1434
+ "grad_norm": 0.004635946121344588,
1435
+ "learning_rate": 0.13884347962746948,
1436
+ "loss": 1.4468,
1437
+ "step": 204
1438
+ },
1439
+ {
1440
+ "epoch": 0.3942307692307692,
1441
+ "grad_norm": 0.00657250913108514,
1442
+ "learning_rate": 0.138268343236509,
1443
+ "loss": 1.4933,
1444
+ "step": 205
1445
+ },
1446
+ {
1447
+ "epoch": 0.39615384615384613,
1448
+ "grad_norm": 0.0044207885444285805,
1449
+ "learning_rate": 0.13769171996210053,
1450
+ "loss": 1.4765,
1451
+ "step": 206
1452
+ },
1453
+ {
1454
+ "epoch": 0.39807692307692305,
1455
+ "grad_norm": 0.004204515533396308,
1456
+ "learning_rate": 0.1371136322084438,
1457
+ "loss": 1.4578,
1458
+ "step": 207
1459
+ },
1460
+ {
1461
+ "epoch": 0.4,
1462
+ "grad_norm": 0.004581208158037667,
1463
+ "learning_rate": 0.13653410243663952,
1464
+ "loss": 1.446,
1465
+ "step": 208
1466
+ },
1467
+ {
1468
+ "epoch": 0.40192307692307694,
1469
+ "grad_norm": 0.0043190870240444965,
1470
+ "learning_rate": 0.13595315316381676,
1471
+ "loss": 1.3652,
1472
+ "step": 209
1473
+ },
1474
+ {
1475
+ "epoch": 0.40384615384615385,
1476
+ "grad_norm": 0.004729893111089475,
1477
+ "learning_rate": 0.13537080696225814,
1478
+ "loss": 1.445,
1479
+ "step": 210
1480
+ },
1481
+ {
1482
+ "epoch": 0.40576923076923077,
1483
+ "grad_norm": 0.004033582633789312,
1484
+ "learning_rate": 0.13478708645852272,
1485
+ "loss": 1.451,
1486
+ "step": 211
1487
+ },
1488
+ {
1489
+ "epoch": 0.4076923076923077,
1490
+ "grad_norm": 0.004249713171656784,
1491
+ "learning_rate": 0.1342020143325669,
1492
+ "loss": 1.4154,
1493
+ "step": 212
1494
+ },
1495
+ {
1496
+ "epoch": 0.4096153846153846,
1497
+ "grad_norm": 0.005314187516060887,
1498
+ "learning_rate": 0.13361561331686309,
1499
+ "loss": 1.3948,
1500
+ "step": 213
1501
+ },
1502
+ {
1503
+ "epoch": 0.4115384615384615,
1504
+ "grad_norm": 0.004534142299321015,
1505
+ "learning_rate": 0.13302790619551672,
1506
+ "loss": 1.379,
1507
+ "step": 214
1508
+ },
1509
+ {
1510
+ "epoch": 0.41346153846153844,
1511
+ "grad_norm": 0.004416951038941333,
1512
+ "learning_rate": 0.1324389158033807,
1513
+ "loss": 1.3812,
1514
+ "step": 215
1515
+ },
1516
+ {
1517
+ "epoch": 0.4153846153846154,
1518
+ "grad_norm": 0.004491041251143997,
1519
+ "learning_rate": 0.13184866502516845,
1520
+ "loss": 1.2882,
1521
+ "step": 216
1522
+ },
1523
+ {
1524
+ "epoch": 0.4173076923076923,
1525
+ "grad_norm": 0.004513666658061728,
1526
+ "learning_rate": 0.13125717679456447,
1527
+ "loss": 1.4169,
1528
+ "step": 217
1529
+ },
1530
+ {
1531
+ "epoch": 0.41923076923076924,
1532
+ "grad_norm": 0.004543959963521464,
1533
+ "learning_rate": 0.13066447409333345,
1534
+ "loss": 1.413,
1535
+ "step": 218
1536
+ },
1537
+ {
1538
+ "epoch": 0.42115384615384616,
1539
+ "grad_norm": 0.00399699603939852,
1540
+ "learning_rate": 0.1300705799504273,
1541
+ "loss": 1.369,
1542
+ "step": 219
1543
+ },
1544
+ {
1545
+ "epoch": 0.4230769230769231,
1546
+ "grad_norm": 0.00609077742566946,
1547
+ "learning_rate": 0.12947551744109043,
1548
+ "loss": 1.4408,
1549
+ "step": 220
1550
+ },
1551
+ {
1552
+ "epoch": 0.425,
1553
+ "grad_norm": 0.004453483715141376,
1554
+ "learning_rate": 0.128879309685963,
1555
+ "loss": 1.4172,
1556
+ "step": 221
1557
+ },
1558
+ {
1559
+ "epoch": 0.4269230769230769,
1560
+ "grad_norm": 0.00477313098283217,
1561
+ "learning_rate": 0.12828197985018275,
1562
+ "loss": 1.3158,
1563
+ "step": 222
1564
+ },
1565
+ {
1566
+ "epoch": 0.4288461538461538,
1567
+ "grad_norm": 0.00410418769542884,
1568
+ "learning_rate": 0.12768355114248495,
1569
+ "loss": 1.3079,
1570
+ "step": 223
1571
+ },
1572
+ {
1573
+ "epoch": 0.4307692307692308,
1574
+ "grad_norm": 0.008269599835992613,
1575
+ "learning_rate": 0.12708404681430052,
1576
+ "loss": 1.6525,
1577
+ "step": 224
1578
+ },
1579
+ {
1580
+ "epoch": 0.4326923076923077,
1581
+ "grad_norm": 0.004979181691672588,
1582
+ "learning_rate": 0.1264834901588527,
1583
+ "loss": 1.3407,
1584
+ "step": 225
1585
+ },
1586
+ {
1587
+ "epoch": 0.4346153846153846,
1588
+ "grad_norm": 0.00424202921173258,
1589
+ "learning_rate": 0.12588190451025208,
1590
+ "loss": 1.4383,
1591
+ "step": 226
1592
+ },
1593
+ {
1594
+ "epoch": 0.43653846153846154,
1595
+ "grad_norm": 0.003962995688081932,
1596
+ "learning_rate": 0.12527931324258976,
1597
+ "loss": 1.4193,
1598
+ "step": 227
1599
+ },
1600
+ {
1601
+ "epoch": 0.43846153846153846,
1602
+ "grad_norm": 0.0051444608986665,
1603
+ "learning_rate": 0.12467573976902935,
1604
+ "loss": 1.5918,
1605
+ "step": 228
1606
+ },
1607
+ {
1608
+ "epoch": 0.4403846153846154,
1609
+ "grad_norm": 0.0038617232746273795,
1610
+ "learning_rate": 0.12407120754089732,
1611
+ "loss": 1.3963,
1612
+ "step": 229
1613
+ },
1614
+ {
1615
+ "epoch": 0.4423076923076923,
1616
+ "grad_norm": 0.004453955857672304,
1617
+ "learning_rate": 0.12346574004677154,
1618
+ "loss": 1.2801,
1619
+ "step": 230
1620
+ },
1621
+ {
1622
+ "epoch": 0.4442307692307692,
1623
+ "grad_norm": 0.004166844708183948,
1624
+ "learning_rate": 0.12285936081156897,
1625
+ "loss": 1.3365,
1626
+ "step": 231
1627
+ },
1628
+ {
1629
+ "epoch": 0.4461538461538462,
1630
+ "grad_norm": 0.0055254304559051405,
1631
+ "learning_rate": 0.12225209339563144,
1632
+ "loss": 1.6206,
1633
+ "step": 232
1634
+ },
1635
+ {
1636
+ "epoch": 0.4480769230769231,
1637
+ "grad_norm": 0.004860072587476447,
1638
+ "learning_rate": 0.12164396139381028,
1639
+ "loss": 1.4937,
1640
+ "step": 233
1641
+ },
1642
+ {
1643
+ "epoch": 0.45,
1644
+ "grad_norm": 0.004387687912717213,
1645
+ "learning_rate": 0.12103498843454959,
1646
+ "loss": 1.2865,
1647
+ "step": 234
1648
+ },
1649
+ {
1650
+ "epoch": 0.4519230769230769,
1651
+ "grad_norm": 0.004961607423103889,
1652
+ "learning_rate": 0.12042519817896805,
1653
+ "loss": 1.3318,
1654
+ "step": 235
1655
+ },
1656
+ {
1657
+ "epoch": 0.45384615384615384,
1658
+ "grad_norm": 0.004246273199153892,
1659
+ "learning_rate": 0.11981461431993977,
1660
+ "loss": 1.4612,
1661
+ "step": 236
1662
+ },
1663
+ {
1664
+ "epoch": 0.45576923076923076,
1665
+ "grad_norm": 0.0042341676045978965,
1666
+ "learning_rate": 0.11920326058117364,
1667
+ "loss": 1.426,
1668
+ "step": 237
1669
+ },
1670
+ {
1671
+ "epoch": 0.4576923076923077,
1672
+ "grad_norm": 0.004464801966764603,
1673
+ "learning_rate": 0.11859116071629149,
1674
+ "loss": 1.3576,
1675
+ "step": 238
1676
+ },
1677
+ {
1678
+ "epoch": 0.4596153846153846,
1679
+ "grad_norm": 0.004674529528799734,
1680
+ "learning_rate": 0.11797833850790528,
1681
+ "loss": 1.466,
1682
+ "step": 239
1683
+ },
1684
+ {
1685
+ "epoch": 0.46153846153846156,
1686
+ "grad_norm": 0.005006829837356061,
1687
+ "learning_rate": 0.11736481776669305,
1688
+ "loss": 1.2198,
1689
+ "step": 240
1690
+ },
1691
+ {
1692
+ "epoch": 0.4634615384615385,
1693
+ "grad_norm": 0.00437058544218353,
1694
+ "learning_rate": 0.11675062233047365,
1695
+ "loss": 1.316,
1696
+ "step": 241
1697
+ },
1698
+ {
1699
+ "epoch": 0.4653846153846154,
1700
+ "grad_norm": 0.0037903834752457807,
1701
+ "learning_rate": 0.11613577606328068,
1702
+ "loss": 1.3297,
1703
+ "step": 242
1704
+ },
1705
+ {
1706
+ "epoch": 0.4673076923076923,
1707
+ "grad_norm": 0.004214200271545134,
1708
+ "learning_rate": 0.11552030285443515,
1709
+ "loss": 1.3213,
1710
+ "step": 243
1711
+ },
1712
+ {
1713
+ "epoch": 0.46923076923076923,
1714
+ "grad_norm": 0.004288624695086091,
1715
+ "learning_rate": 0.11490422661761744,
1716
+ "loss": 1.4657,
1717
+ "step": 244
1718
+ },
1719
+ {
1720
+ "epoch": 0.47115384615384615,
1721
+ "grad_norm": 0.00439225457121165,
1722
+ "learning_rate": 0.11428757128993801,
1723
+ "loss": 1.3135,
1724
+ "step": 245
1725
+ },
1726
+ {
1727
+ "epoch": 0.47307692307692306,
1728
+ "grad_norm": 0.004574255544147889,
1729
+ "learning_rate": 0.11367036083100734,
1730
+ "loss": 1.5876,
1731
+ "step": 246
1732
+ },
1733
+ {
1734
+ "epoch": 0.475,
1735
+ "grad_norm": 0.004138239283667433,
1736
+ "learning_rate": 0.11305261922200519,
1737
+ "loss": 1.484,
1738
+ "step": 247
1739
+ },
1740
+ {
1741
+ "epoch": 0.47692307692307695,
1742
+ "grad_norm": 0.004146625545960324,
1743
+ "learning_rate": 0.11243437046474854,
1744
+ "loss": 1.3073,
1745
+ "step": 248
1746
+ },
1747
+ {
1748
+ "epoch": 0.47884615384615387,
1749
+ "grad_norm": 0.004691074032038318,
1750
+ "learning_rate": 0.1118156385807593,
1751
+ "loss": 1.4229,
1752
+ "step": 249
1753
+ },
1754
+ {
1755
+ "epoch": 0.4807692307692308,
1756
+ "grad_norm": 0.005345229808210592,
1757
+ "learning_rate": 0.11119644761033079,
1758
+ "loss": 1.3579,
1759
+ "step": 250
1760
+ },
1761
+ {
1762
+ "epoch": 0.4826923076923077,
1763
+ "grad_norm": 0.00380833494041057,
1764
+ "learning_rate": 0.1105768216115938,
1765
+ "loss": 1.4183,
1766
+ "step": 251
1767
+ },
1768
+ {
1769
+ "epoch": 0.4846153846153846,
1770
+ "grad_norm": 0.004142663741123374,
1771
+ "learning_rate": 0.10995678465958168,
1772
+ "loss": 1.456,
1773
+ "step": 252
1774
+ },
1775
+ {
1776
+ "epoch": 0.48653846153846153,
1777
+ "grad_norm": 0.00475522896067983,
1778
+ "learning_rate": 0.10933636084529506,
1779
+ "loss": 1.4186,
1780
+ "step": 253
1781
+ },
1782
+ {
1783
+ "epoch": 0.48846153846153845,
1784
+ "grad_norm": 0.003876598601354388,
1785
+ "learning_rate": 0.10871557427476584,
1786
+ "loss": 1.3254,
1787
+ "step": 254
1788
+ },
1789
+ {
1790
+ "epoch": 0.49038461538461536,
1791
+ "grad_norm": 0.004379447504271326,
1792
+ "learning_rate": 0.10809444906812034,
1793
+ "loss": 1.339,
1794
+ "step": 255
1795
+ },
1796
+ {
1797
+ "epoch": 0.49230769230769234,
1798
+ "grad_norm": 0.004630489830219677,
1799
+ "learning_rate": 0.10747300935864244,
1800
+ "loss": 1.3829,
1801
+ "step": 256
1802
+ },
1803
+ {
1804
+ "epoch": 0.49423076923076925,
1805
+ "grad_norm": 0.0041589182482135185,
1806
+ "learning_rate": 0.10685127929183567,
1807
+ "loss": 1.3781,
1808
+ "step": 257
1809
+ },
1810
+ {
1811
+ "epoch": 0.49615384615384617,
1812
+ "grad_norm": 0.003622975625798608,
1813
+ "learning_rate": 0.10622928302448523,
1814
+ "loss": 1.3963,
1815
+ "step": 258
1816
+ },
1817
+ {
1818
+ "epoch": 0.4980769230769231,
1819
+ "grad_norm": 0.004764173350396567,
1820
+ "learning_rate": 0.10560704472371918,
1821
+ "loss": 1.4706,
1822
+ "step": 259
1823
+ },
1824
+ {
1825
+ "epoch": 0.5,
1826
+ "grad_norm": 0.004051179827119963,
1827
+ "learning_rate": 0.10498458856606972,
1828
+ "loss": 1.5381,
1829
+ "step": 260
1830
+ },
1831
+ {
1832
+ "epoch": 0.5019230769230769,
1833
+ "grad_norm": 0.00414587804797381,
1834
+ "learning_rate": 0.10436193873653361,
1835
+ "loss": 1.4645,
1836
+ "step": 261
1837
+ },
1838
+ {
1839
+ "epoch": 0.5038461538461538,
1840
+ "grad_norm": 0.004355455276573701,
1841
+ "learning_rate": 0.10373911942763259,
1842
+ "loss": 1.3089,
1843
+ "step": 262
1844
+ },
1845
+ {
1846
+ "epoch": 0.5057692307692307,
1847
+ "grad_norm": 0.0046757438980203556,
1848
+ "learning_rate": 0.10311615483847332,
1849
+ "loss": 1.4706,
1850
+ "step": 263
1851
+ },
1852
+ {
1853
+ "epoch": 0.5076923076923077,
1854
+ "grad_norm": 0.003998988939470536,
1855
+ "learning_rate": 0.1024930691738073,
1856
+ "loss": 1.4215,
1857
+ "step": 264
1858
+ },
1859
+ {
1860
+ "epoch": 0.5096153846153846,
1861
+ "grad_norm": 0.005392604182356352,
1862
+ "learning_rate": 0.10186988664309023,
1863
+ "loss": 1.3282,
1864
+ "step": 265
1865
+ },
1866
+ {
1867
+ "epoch": 0.5115384615384615,
1868
+ "grad_norm": 0.0035569159546779915,
1869
+ "learning_rate": 0.10124663145954152,
1870
+ "loss": 1.1636,
1871
+ "step": 266
1872
+ },
1873
+ {
1874
+ "epoch": 0.5134615384615384,
1875
+ "grad_norm": 0.003837956206262098,
1876
+ "learning_rate": 0.10062332783920336,
1877
+ "loss": 1.3125,
1878
+ "step": 267
1879
+ },
1880
+ {
1881
+ "epoch": 0.5153846153846153,
1882
+ "grad_norm": 0.005782906690951889,
1883
+ "learning_rate": 0.1,
1884
+ "loss": 1.6067,
1885
+ "step": 268
1886
+ },
1887
+ {
1888
+ "epoch": 0.5173076923076924,
1889
+ "grad_norm": 0.004381502781854409,
1890
+ "learning_rate": 0.09937667216079665,
1891
+ "loss": 1.4352,
1892
+ "step": 269
1893
+ },
1894
+ {
1895
+ "epoch": 0.5192307692307693,
1896
+ "grad_norm": 0.003998861140428057,
1897
+ "learning_rate": 0.0987533685404585,
1898
+ "loss": 1.3802,
1899
+ "step": 270
1900
+ },
1901
+ {
1902
+ "epoch": 0.5211538461538462,
1903
+ "grad_norm": 0.004236288934399927,
1904
+ "learning_rate": 0.0981301133569098,
1905
+ "loss": 1.417,
1906
+ "step": 271
1907
+ },
1908
+ {
1909
+ "epoch": 0.5230769230769231,
1910
+ "grad_norm": 0.00491273282387985,
1911
+ "learning_rate": 0.09750693082619273,
1912
+ "loss": 1.397,
1913
+ "step": 272
1914
+ },
1915
+ {
1916
+ "epoch": 0.525,
1917
+ "grad_norm": 0.004798064203901598,
1918
+ "learning_rate": 0.0968838451615267,
1919
+ "loss": 1.5385,
1920
+ "step": 273
1921
+ },
1922
+ {
1923
+ "epoch": 0.5269230769230769,
1924
+ "grad_norm": 0.004478156309093749,
1925
+ "learning_rate": 0.09626088057236745,
1926
+ "loss": 1.3716,
1927
+ "step": 274
1928
+ },
1929
+ {
1930
+ "epoch": 0.5288461538461539,
1931
+ "grad_norm": 0.004826756385375406,
1932
+ "learning_rate": 0.09563806126346641,
1933
+ "loss": 1.31,
1934
+ "step": 275
1935
+ },
1936
+ {
1937
+ "epoch": 0.5307692307692308,
1938
+ "grad_norm": 0.004458046390301229,
1939
+ "learning_rate": 0.09501541143393027,
1940
+ "loss": 1.402,
1941
+ "step": 276
1942
+ },
1943
+ {
1944
+ "epoch": 0.5326923076923077,
1945
+ "grad_norm": 0.005353963724542705,
1946
+ "learning_rate": 0.09439295527628082,
1947
+ "loss": 1.5191,
1948
+ "step": 277
1949
+ },
1950
+ {
1951
+ "epoch": 0.5346153846153846,
1952
+ "grad_norm": 0.0038307163020152167,
1953
+ "learning_rate": 0.0937707169755148,
1954
+ "loss": 1.2593,
1955
+ "step": 278
1956
+ },
1957
+ {
1958
+ "epoch": 0.5365384615384615,
1959
+ "grad_norm": 0.006456696602144017,
1960
+ "learning_rate": 0.09314872070816434,
1961
+ "loss": 1.4252,
1962
+ "step": 279
1963
+ },
1964
+ {
1965
+ "epoch": 0.5384615384615384,
1966
+ "grad_norm": 0.004880749144261153,
1967
+ "learning_rate": 0.09252699064135758,
1968
+ "loss": 1.3187,
1969
+ "step": 280
1970
+ },
1971
+ {
1972
+ "epoch": 0.5403846153846154,
1973
+ "grad_norm": 0.004470796506295151,
1974
+ "learning_rate": 0.09190555093187967,
1975
+ "loss": 1.4278,
1976
+ "step": 281
1977
+ },
1978
+ {
1979
+ "epoch": 0.5423076923076923,
1980
+ "grad_norm": 0.0040297815554439335,
1981
+ "learning_rate": 0.09128442572523418,
1982
+ "loss": 1.2662,
1983
+ "step": 282
1984
+ },
1985
+ {
1986
+ "epoch": 0.5442307692307692,
1987
+ "grad_norm": 0.004922877811429402,
1988
+ "learning_rate": 0.09066363915470495,
1989
+ "loss": 1.4582,
1990
+ "step": 283
1991
+ },
1992
+ {
1993
+ "epoch": 0.5461538461538461,
1994
+ "grad_norm": 0.005228969551421608,
1995
+ "learning_rate": 0.09004321534041836,
1996
+ "loss": 1.388,
1997
+ "step": 284
1998
+ },
1999
+ {
2000
+ "epoch": 0.5480769230769231,
2001
+ "grad_norm": 0.004356044094551331,
2002
+ "learning_rate": 0.08942317838840624,
2003
+ "loss": 1.3011,
2004
+ "step": 285
2005
+ },
2006
+ {
2007
+ "epoch": 0.55,
2008
+ "grad_norm": 0.00458273147690172,
2009
+ "learning_rate": 0.08880355238966922,
2010
+ "loss": 1.1633,
2011
+ "step": 286
2012
+ },
2013
+ {
2014
+ "epoch": 0.551923076923077,
2015
+ "grad_norm": 0.004485798161907241,
2016
+ "learning_rate": 0.08818436141924073,
2017
+ "loss": 1.4253,
2018
+ "step": 287
2019
+ },
2020
+ {
2021
+ "epoch": 0.5538461538461539,
2022
+ "grad_norm": 0.004737845940265029,
2023
+ "learning_rate": 0.08756562953525152,
2024
+ "loss": 1.4706,
2025
+ "step": 288
2026
+ },
2027
+ {
2028
+ "epoch": 0.5557692307692308,
2029
+ "grad_norm": 0.004028175288879505,
2030
+ "learning_rate": 0.08694738077799487,
2031
+ "loss": 1.3182,
2032
+ "step": 289
2033
+ },
2034
+ {
2035
+ "epoch": 0.5576923076923077,
2036
+ "grad_norm": 0.00378878204635607,
2037
+ "learning_rate": 0.08632963916899268,
2038
+ "loss": 1.2282,
2039
+ "step": 290
2040
+ },
2041
+ {
2042
+ "epoch": 0.5596153846153846,
2043
+ "grad_norm": 0.00474752974281256,
2044
+ "learning_rate": 0.08571242871006202,
2045
+ "loss": 1.2921,
2046
+ "step": 291
2047
+ },
2048
+ {
2049
+ "epoch": 0.5615384615384615,
2050
+ "grad_norm": 0.00408336026753124,
2051
+ "learning_rate": 0.08509577338238256,
2052
+ "loss": 1.3415,
2053
+ "step": 292
2054
+ },
2055
+ {
2056
+ "epoch": 0.5634615384615385,
2057
+ "grad_norm": 0.004484080280604913,
2058
+ "learning_rate": 0.08447969714556484,
2059
+ "loss": 1.2751,
2060
+ "step": 293
2061
+ },
2062
+ {
2063
+ "epoch": 0.5653846153846154,
2064
+ "grad_norm": 0.004341379289954433,
2065
+ "learning_rate": 0.08386422393671933,
2066
+ "loss": 1.3136,
2067
+ "step": 294
2068
+ },
2069
+ {
2070
+ "epoch": 0.5673076923076923,
2071
+ "grad_norm": 0.005634165425607994,
2072
+ "learning_rate": 0.08324937766952638,
2073
+ "loss": 1.4681,
2074
+ "step": 295
2075
+ },
2076
+ {
2077
+ "epoch": 0.5692307692307692,
2078
+ "grad_norm": 0.004459802187914876,
2079
+ "learning_rate": 0.08263518223330697,
2080
+ "loss": 1.2474,
2081
+ "step": 296
2082
+ },
2083
+ {
2084
+ "epoch": 0.5711538461538461,
2085
+ "grad_norm": 0.004561691362059065,
2086
+ "learning_rate": 0.08202166149209474,
2087
+ "loss": 1.3937,
2088
+ "step": 297
2089
+ },
2090
+ {
2091
+ "epoch": 0.573076923076923,
2092
+ "grad_norm": 0.0038830636098779845,
2093
+ "learning_rate": 0.08140883928370855,
2094
+ "loss": 1.3523,
2095
+ "step": 298
2096
+ },
2097
+ {
2098
+ "epoch": 0.575,
2099
+ "grad_norm": 0.004481460408952819,
2100
+ "learning_rate": 0.0807967394188264,
2101
+ "loss": 1.464,
2102
+ "step": 299
2103
+ },
2104
+ {
2105
+ "epoch": 0.5769230769230769,
2106
+ "grad_norm": 0.004307253957155686,
2107
+ "learning_rate": 0.08018538568006027,
2108
+ "loss": 1.4132,
2109
+ "step": 300
2110
+ },
2111
+ {
2112
+ "epoch": 0.5788461538461539,
2113
+ "grad_norm": 0.0041670150767262795,
2114
+ "learning_rate": 0.07957480182103199,
2115
+ "loss": 1.3816,
2116
+ "step": 301
2117
+ },
2118
+ {
2119
+ "epoch": 0.5807692307692308,
2120
+ "grad_norm": 0.004538554443036844,
2121
+ "learning_rate": 0.07896501156545044,
2122
+ "loss": 1.4821,
2123
+ "step": 302
2124
+ },
2125
+ {
2126
+ "epoch": 0.5826923076923077,
2127
+ "grad_norm": 0.004330197551507648,
2128
+ "learning_rate": 0.07835603860618973,
2129
+ "loss": 1.3102,
2130
+ "step": 303
2131
+ },
2132
+ {
2133
+ "epoch": 0.5846153846153846,
2134
+ "grad_norm": 0.004378184160937378,
2135
+ "learning_rate": 0.07774790660436857,
2136
+ "loss": 1.3757,
2137
+ "step": 304
2138
+ },
2139
+ {
2140
+ "epoch": 0.5865384615384616,
2141
+ "grad_norm": 0.0047027419446543026,
2142
+ "learning_rate": 0.07714063918843106,
2143
+ "loss": 1.4328,
2144
+ "step": 305
2145
+ },
2146
+ {
2147
+ "epoch": 0.5884615384615385,
2148
+ "grad_norm": 0.004135029170448559,
2149
+ "learning_rate": 0.0765342599532285,
2150
+ "loss": 1.3621,
2151
+ "step": 306
2152
+ },
2153
+ {
2154
+ "epoch": 0.5903846153846154,
2155
+ "grad_norm": 0.004428193454528292,
2156
+ "learning_rate": 0.07592879245910272,
2157
+ "loss": 1.3021,
2158
+ "step": 307
2159
+ },
2160
+ {
2161
+ "epoch": 0.5923076923076923,
2162
+ "grad_norm": 0.004111361321866392,
2163
+ "learning_rate": 0.07532426023097064,
2164
+ "loss": 1.4195,
2165
+ "step": 308
2166
+ },
2167
+ {
2168
+ "epoch": 0.5942307692307692,
2169
+ "grad_norm": 0.003934055801317284,
2170
+ "learning_rate": 0.07472068675741024,
2171
+ "loss": 1.2865,
2172
+ "step": 309
2173
+ },
2174
+ {
2175
+ "epoch": 0.5961538461538461,
2176
+ "grad_norm": 0.004012945545338028,
2177
+ "learning_rate": 0.07411809548974792,
2178
+ "loss": 1.264,
2179
+ "step": 310
2180
+ },
2181
+ {
2182
+ "epoch": 0.5980769230769231,
2183
+ "grad_norm": 0.004091237474728546,
2184
+ "learning_rate": 0.07351650984114727,
2185
+ "loss": 1.2307,
2186
+ "step": 311
2187
+ },
2188
+ {
2189
+ "epoch": 0.6,
2190
+ "grad_norm": 0.00522677951723774,
2191
+ "learning_rate": 0.0729159531856995,
2192
+ "loss": 1.2313,
2193
+ "step": 312
2194
+ },
2195
+ {
2196
+ "epoch": 0.6019230769230769,
2197
+ "grad_norm": 0.0038247111991477414,
2198
+ "learning_rate": 0.07231644885751508,
2199
+ "loss": 1.2184,
2200
+ "step": 313
2201
+ },
2202
+ {
2203
+ "epoch": 0.6038461538461538,
2204
+ "grad_norm": 0.003826034149415262,
2205
+ "learning_rate": 0.07171802014981725,
2206
+ "loss": 1.2567,
2207
+ "step": 314
2208
+ },
2209
+ {
2210
+ "epoch": 0.6057692307692307,
2211
+ "grad_norm": 0.006763985313927637,
2212
+ "learning_rate": 0.07112069031403703,
2213
+ "loss": 1.4729,
2214
+ "step": 315
2215
+ },
2216
+ {
2217
+ "epoch": 0.6076923076923076,
2218
+ "grad_norm": 0.0049476497086911745,
2219
+ "learning_rate": 0.07052448255890957,
2220
+ "loss": 1.2331,
2221
+ "step": 316
2222
+ },
2223
+ {
2224
+ "epoch": 0.6096153846153847,
2225
+ "grad_norm": 0.003871856329153639,
2226
+ "learning_rate": 0.0699294200495727,
2227
+ "loss": 1.2506,
2228
+ "step": 317
2229
+ },
2230
+ {
2231
+ "epoch": 0.6115384615384616,
2232
+ "grad_norm": 0.004399092771163042,
2233
+ "learning_rate": 0.06933552590666658,
2234
+ "loss": 1.3806,
2235
+ "step": 318
2236
+ },
2237
+ {
2238
+ "epoch": 0.6134615384615385,
2239
+ "grad_norm": 0.004145275768242621,
2240
+ "learning_rate": 0.06874282320543557,
2241
+ "loss": 1.239,
2242
+ "step": 319
2243
+ },
2244
+ {
2245
+ "epoch": 0.6153846153846154,
2246
+ "grad_norm": 0.004378764090972851,
2247
+ "learning_rate": 0.06815133497483157,
2248
+ "loss": 1.1812,
2249
+ "step": 320
2250
+ },
2251
+ {
2252
+ "epoch": 0.6173076923076923,
2253
+ "grad_norm": 0.004224448902803425,
2254
+ "learning_rate": 0.06756108419661931,
2255
+ "loss": 1.3886,
2256
+ "step": 321
2257
+ },
2258
+ {
2259
+ "epoch": 0.6192307692307693,
2260
+ "grad_norm": 0.0044144089463406,
2261
+ "learning_rate": 0.06697209380448332,
2262
+ "loss": 1.2948,
2263
+ "step": 322
2264
+ },
2265
+ {
2266
+ "epoch": 0.6211538461538462,
2267
+ "grad_norm": 0.005978835240941658,
2268
+ "learning_rate": 0.06638438668313694,
2269
+ "loss": 1.3875,
2270
+ "step": 323
2271
+ },
2272
+ {
2273
+ "epoch": 0.6230769230769231,
2274
+ "grad_norm": 0.00482129421919115,
2275
+ "learning_rate": 0.06579798566743314,
2276
+ "loss": 1.3141,
2277
+ "step": 324
2278
+ },
2279
+ {
2280
+ "epoch": 0.625,
2281
+ "grad_norm": 0.004338837790268412,
2282
+ "learning_rate": 0.06521291354147728,
2283
+ "loss": 1.3307,
2284
+ "step": 325
2285
+ },
2286
+ {
2287
+ "epoch": 0.6269230769230769,
2288
+ "grad_norm": 0.004316085641884237,
2289
+ "learning_rate": 0.06462919303774187,
2290
+ "loss": 1.3107,
2291
+ "step": 326
2292
+ },
2293
+ {
2294
+ "epoch": 0.6288461538461538,
2295
+ "grad_norm": 0.006014611251188303,
2296
+ "learning_rate": 0.06404684683618325,
2297
+ "loss": 1.4589,
2298
+ "step": 327
2299
+ },
2300
+ {
2301
+ "epoch": 0.6307692307692307,
2302
+ "grad_norm": 0.004292936777259576,
2303
+ "learning_rate": 0.0634658975633605,
2304
+ "loss": 1.3884,
2305
+ "step": 328
2306
+ },
2307
+ {
2308
+ "epoch": 0.6326923076923077,
2309
+ "grad_norm": 0.003659574102568691,
2310
+ "learning_rate": 0.06288636779155621,
2311
+ "loss": 1.2259,
2312
+ "step": 329
2313
+ },
2314
+ {
2315
+ "epoch": 0.6346153846153846,
2316
+ "grad_norm": 0.0037607700376099964,
2317
+ "learning_rate": 0.06230828003789948,
2318
+ "loss": 1.3025,
2319
+ "step": 330
2320
+ },
2321
+ {
2322
+ "epoch": 0.6365384615384615,
2323
+ "grad_norm": 0.004018380336053225,
2324
+ "learning_rate": 0.06173165676349103,
2325
+ "loss": 1.2716,
2326
+ "step": 331
2327
+ },
2328
+ {
2329
+ "epoch": 0.6384615384615384,
2330
+ "grad_norm": 0.004064795838143721,
2331
+ "learning_rate": 0.06115652037253053,
2332
+ "loss": 1.4522,
2333
+ "step": 332
2334
+ },
2335
+ {
2336
+ "epoch": 0.6403846153846153,
2337
+ "grad_norm": 0.004378231690112899,
2338
+ "learning_rate": 0.06058289321144608,
2339
+ "loss": 1.4485,
2340
+ "step": 333
2341
+ },
2342
+ {
2343
+ "epoch": 0.6423076923076924,
2344
+ "grad_norm": 0.004691962067907288,
2345
+ "learning_rate": 0.06001079756802592,
2346
+ "loss": 1.3259,
2347
+ "step": 334
2348
+ },
2349
+ {
2350
+ "epoch": 0.6442307692307693,
2351
+ "grad_norm": 0.004045415295751982,
2352
+ "learning_rate": 0.059440255670552514,
2353
+ "loss": 1.3137,
2354
+ "step": 335
2355
+ },
2356
+ {
2357
+ "epoch": 0.6461538461538462,
2358
+ "grad_norm": 0.004520510422636476,
2359
+ "learning_rate": 0.05887128968693887,
2360
+ "loss": 1.1934,
2361
+ "step": 336
2362
+ },
2363
+ {
2364
+ "epoch": 0.6480769230769231,
2365
+ "grad_norm": 0.004389249477802447,
2366
+ "learning_rate": 0.058303921723867225,
2367
+ "loss": 1.1964,
2368
+ "step": 337
2369
+ },
2370
+ {
2371
+ "epoch": 0.65,
2372
+ "grad_norm": 0.004284427231829871,
2373
+ "learning_rate": 0.05773817382593008,
2374
+ "loss": 1.3311,
2375
+ "step": 338
2376
+ },
2377
+ {
2378
+ "epoch": 0.6519230769230769,
2379
+ "grad_norm": 0.004299084924868999,
2380
+ "learning_rate": 0.057174067974773715,
2381
+ "loss": 1.2661,
2382
+ "step": 339
2383
+ },
2384
+ {
2385
+ "epoch": 0.6538461538461539,
2386
+ "grad_norm": 0.004142944351678658,
2387
+ "learning_rate": 0.056611626088244195,
2388
+ "loss": 1.2505,
2389
+ "step": 340
2390
+ },
2391
+ {
2392
+ "epoch": 0.6557692307692308,
2393
+ "grad_norm": 0.004273001392782499,
2394
+ "learning_rate": 0.056050870019535494,
2395
+ "loss": 1.277,
2396
+ "step": 341
2397
+ },
2398
+ {
2399
+ "epoch": 0.6576923076923077,
2400
+ "grad_norm": 0.004890174107219532,
2401
+ "learning_rate": 0.05549182155634076,
2402
+ "loss": 1.4282,
2403
+ "step": 342
2404
+ },
2405
+ {
2406
+ "epoch": 0.6596153846153846,
2407
+ "grad_norm": 0.00465483001229315,
2408
+ "learning_rate": 0.054934502420005464,
2409
+ "loss": 1.3933,
2410
+ "step": 343
2411
+ },
2412
+ {
2413
+ "epoch": 0.6615384615384615,
2414
+ "grad_norm": 0.003975560170268734,
2415
+ "learning_rate": 0.0543789342646837,
2416
+ "loss": 1.2218,
2417
+ "step": 344
2418
+ },
2419
+ {
2420
+ "epoch": 0.6634615384615384,
2421
+ "grad_norm": 0.004520470968366567,
2422
+ "learning_rate": 0.05382513867649663,
2423
+ "loss": 1.3484,
2424
+ "step": 345
2425
+ },
2426
+ {
2427
+ "epoch": 0.6653846153846154,
2428
+ "grad_norm": 0.004375057258916997,
2429
+ "learning_rate": 0.0532731371726938,
2430
+ "loss": 1.3758,
2431
+ "step": 346
2432
+ },
2433
+ {
2434
+ "epoch": 0.6673076923076923,
2435
+ "grad_norm": 0.003909849226708089,
2436
+ "learning_rate": 0.05272295120081732,
2437
+ "loss": 1.2463,
2438
+ "step": 347
2439
+ },
2440
+ {
2441
+ "epoch": 0.6692307692307692,
2442
+ "grad_norm": 0.004728265597326192,
2443
+ "learning_rate": 0.05217460213786821,
2444
+ "loss": 1.2027,
2445
+ "step": 348
2446
+ },
2447
+ {
2448
+ "epoch": 0.6711538461538461,
2449
+ "grad_norm": 0.004454076142100732,
2450
+ "learning_rate": 0.051628111289476024,
2451
+ "loss": 1.255,
2452
+ "step": 349
2453
+ },
2454
+ {
2455
+ "epoch": 0.6730769230769231,
2456
+ "grad_norm": 0.004475124925757971,
2457
+ "learning_rate": 0.051083499889071106,
2458
+ "loss": 1.2805,
2459
+ "step": 350
2460
+ },
2461
+ {
2462
+ "epoch": 0.675,
2463
+ "grad_norm": 0.003964419403654661,
2464
+ "learning_rate": 0.05054078909705926,
2465
+ "loss": 1.1869,
2466
+ "step": 351
2467
+ },
2468
+ {
2469
+ "epoch": 0.676923076923077,
2470
+ "grad_norm": 0.004000264982003876,
2471
+ "learning_rate": 0.050000000000000024,
2472
+ "loss": 1.3151,
2473
+ "step": 352
2474
+ },
2475
+ {
2476
+ "epoch": 0.6788461538461539,
2477
+ "grad_norm": 0.003737847077837895,
2478
+ "learning_rate": 0.04946115360978696,
2479
+ "loss": 1.3195,
2480
+ "step": 353
2481
+ },
2482
+ {
2483
+ "epoch": 0.6807692307692308,
2484
+ "grad_norm": 0.004370826907455206,
2485
+ "learning_rate": 0.048924270862831465,
2486
+ "loss": 1.4626,
2487
+ "step": 354
2488
+ },
2489
+ {
2490
+ "epoch": 0.6826923076923077,
2491
+ "grad_norm": 0.003991108589508579,
2492
+ "learning_rate": 0.04838937261924933,
2493
+ "loss": 1.2531,
2494
+ "step": 355
2495
+ },
2496
+ {
2497
+ "epoch": 0.6846153846153846,
2498
+ "grad_norm": 0.004017476931403349,
2499
+ "learning_rate": 0.0478564796620502,
2500
+ "loss": 1.2502,
2501
+ "step": 356
2502
+ },
2503
+ {
2504
+ "epoch": 0.6865384615384615,
2505
+ "grad_norm": 0.0035983509929755284,
2506
+ "learning_rate": 0.04732561269632993,
2507
+ "loss": 1.2686,
2508
+ "step": 357
2509
+ },
2510
+ {
2511
+ "epoch": 0.6884615384615385,
2512
+ "grad_norm": 0.004011748365752871,
2513
+ "learning_rate": 0.04679679234846636,
2514
+ "loss": 1.2043,
2515
+ "step": 358
2516
+ },
2517
+ {
2518
+ "epoch": 0.6903846153846154,
2519
+ "grad_norm": 0.0044624363898576715,
2520
+ "learning_rate": 0.046270039165317606,
2521
+ "loss": 1.3817,
2522
+ "step": 359
2523
+ },
2524
+ {
2525
+ "epoch": 0.6923076923076923,
2526
+ "grad_norm": 0.005344574436184053,
2527
+ "learning_rate": 0.04574537361342407,
2528
+ "loss": 1.4085,
2529
+ "step": 360
2530
+ },
2531
+ {
2532
+ "epoch": 0.6942307692307692,
2533
+ "grad_norm": 0.004039467262017942,
2534
+ "learning_rate": 0.04522281607821288,
2535
+ "loss": 1.3928,
2536
+ "step": 361
2537
+ },
2538
+ {
2539
+ "epoch": 0.6961538461538461,
2540
+ "grad_norm": 0.004361818655059272,
2541
+ "learning_rate": 0.04470238686320606,
2542
+ "loss": 1.2693,
2543
+ "step": 362
2544
+ },
2545
+ {
2546
+ "epoch": 0.698076923076923,
2547
+ "grad_norm": 0.0040109714384689585,
2548
+ "learning_rate": 0.044184106189231624,
2549
+ "loss": 1.3009,
2550
+ "step": 363
2551
+ },
2552
+ {
2553
+ "epoch": 0.7,
2554
+ "grad_norm": 0.0041728376345650025,
2555
+ "learning_rate": 0.043667994193637795,
2556
+ "loss": 1.4058,
2557
+ "step": 364
2558
+ },
2559
+ {
2560
+ "epoch": 0.7019230769230769,
2561
+ "grad_norm": 0.004282868380076848,
2562
+ "learning_rate": 0.043154070929510784,
2563
+ "loss": 1.3641,
2564
+ "step": 365
2565
+ },
2566
+ {
2567
+ "epoch": 0.7038461538461539,
2568
+ "grad_norm": 0.003926864120183499,
2569
+ "learning_rate": 0.04264235636489542,
2570
+ "loss": 1.3109,
2571
+ "step": 366
2572
+ },
2573
+ {
2574
+ "epoch": 0.7057692307692308,
2575
+ "grad_norm": 0.003979261584259288,
2576
+ "learning_rate": 0.04213287038201943,
2577
+ "loss": 1.3143,
2578
+ "step": 367
2579
+ },
2580
+ {
2581
+ "epoch": 0.7076923076923077,
2582
+ "grad_norm": 0.004421281737764511,
2583
+ "learning_rate": 0.04162563277652104,
2584
+ "loss": 1.1646,
2585
+ "step": 368
2586
+ },
2587
+ {
2588
+ "epoch": 0.7096153846153846,
2589
+ "grad_norm": 0.004385249875479428,
2590
+ "learning_rate": 0.04112066325667954,
2591
+ "loss": 1.3621,
2592
+ "step": 369
2593
+ },
2594
+ {
2595
+ "epoch": 0.7115384615384616,
2596
+ "grad_norm": 0.0036803497737187825,
2597
+ "learning_rate": 0.04061798144264986,
2598
+ "loss": 1.2177,
2599
+ "step": 370
2600
+ },
2601
+ {
2602
+ "epoch": 0.7134615384615385,
2603
+ "grad_norm": 0.004146459162710883,
2604
+ "learning_rate": 0.04011760686569998,
2605
+ "loss": 1.215,
2606
+ "step": 371
2607
+ },
2608
+ {
2609
+ "epoch": 0.7153846153846154,
2610
+ "grad_norm": 0.0038958662345668416,
2611
+ "learning_rate": 0.03961955896745224,
2612
+ "loss": 1.46,
2613
+ "step": 372
2614
+ },
2615
+ {
2616
+ "epoch": 0.7173076923076923,
2617
+ "grad_norm": 0.004525706798703796,
2618
+ "learning_rate": 0.03912385709912794,
2619
+ "loss": 1.3312,
2620
+ "step": 373
2621
+ },
2622
+ {
2623
+ "epoch": 0.7192307692307692,
2624
+ "grad_norm": 0.004196672738286894,
2625
+ "learning_rate": 0.038630520520795276,
2626
+ "loss": 1.2991,
2627
+ "step": 374
2628
+ },
2629
+ {
2630
+ "epoch": 0.7211538461538461,
2631
+ "grad_norm": 0.0038652167667370574,
2632
+ "learning_rate": 0.03813956840062119,
2633
+ "loss": 1.2072,
2634
+ "step": 375
2635
+ },
2636
+ {
2637
+ "epoch": 0.7230769230769231,
2638
+ "grad_norm": 0.003827130692066178,
2639
+ "learning_rate": 0.037651019814126656,
2640
+ "loss": 1.3295,
2641
+ "step": 376
2642
+ },
2643
+ {
2644
+ "epoch": 0.725,
2645
+ "grad_norm": 0.007301757829963305,
2646
+ "learning_rate": 0.037164893743445275,
2647
+ "loss": 1.2701,
2648
+ "step": 377
2649
+ },
2650
+ {
2651
+ "epoch": 0.7269230769230769,
2652
+ "grad_norm": 0.003945477387280622,
2653
+ "learning_rate": 0.03668120907658603,
2654
+ "loss": 1.3215,
2655
+ "step": 378
2656
+ },
2657
+ {
2658
+ "epoch": 0.7288461538461538,
2659
+ "grad_norm": 0.003928540009880384,
2660
+ "learning_rate": 0.036199984606699154,
2661
+ "loss": 1.3061,
2662
+ "step": 379
2663
+ },
2664
+ {
2665
+ "epoch": 0.7307692307692307,
2666
+ "grad_norm": 0.0053992479123445335,
2667
+ "learning_rate": 0.035721239031346066,
2668
+ "loss": 1.4423,
2669
+ "step": 380
2670
+ },
2671
+ {
2672
+ "epoch": 0.7326923076923076,
2673
+ "grad_norm": 0.004062662921463541,
2674
+ "learning_rate": 0.03524499095177297,
2675
+ "loss": 1.2957,
2676
+ "step": 381
2677
+ },
2678
+ {
2679
+ "epoch": 0.7346153846153847,
2680
+ "grad_norm": 0.004207519927919408,
2681
+ "learning_rate": 0.03477125887218792,
2682
+ "loss": 1.3756,
2683
+ "step": 382
2684
+ },
2685
+ {
2686
+ "epoch": 0.7365384615384616,
2687
+ "grad_norm": 0.004526323637346153,
2688
+ "learning_rate": 0.03430006119904196,
2689
+ "loss": 1.1317,
2690
+ "step": 383
2691
+ },
2692
+ {
2693
+ "epoch": 0.7384615384615385,
2694
+ "grad_norm": 0.004473271335535201,
2695
+ "learning_rate": 0.033831416240314084,
2696
+ "loss": 1.4992,
2697
+ "step": 384
2698
+ },
2699
+ {
2700
+ "epoch": 0.7403846153846154,
2701
+ "grad_norm": 0.0037578159533615324,
2702
+ "learning_rate": 0.03336534220479961,
2703
+ "loss": 1.2748,
2704
+ "step": 385
2705
+ },
2706
+ {
2707
+ "epoch": 0.7423076923076923,
2708
+ "grad_norm": 0.003601830206175373,
2709
+ "learning_rate": 0.032901857201403005,
2710
+ "loss": 1.2257,
2711
+ "step": 386
2712
+ },
2713
+ {
2714
+ "epoch": 0.7442307692307693,
2715
+ "grad_norm": 0.004239088032869379,
2716
+ "learning_rate": 0.032440979238433976,
2717
+ "loss": 1.4613,
2718
+ "step": 387
2719
+ },
2720
+ {
2721
+ "epoch": 0.7461538461538462,
2722
+ "grad_norm": 0.0037790262720102173,
2723
+ "learning_rate": 0.03198272622290804,
2724
+ "loss": 1.1697,
2725
+ "step": 388
2726
+ },
2727
+ {
2728
+ "epoch": 0.7480769230769231,
2729
+ "grad_norm": 0.004775229669855552,
2730
+ "learning_rate": 0.03152711595985065,
2731
+ "loss": 1.236,
2732
+ "step": 389
2733
+ },
2734
+ {
2735
+ "epoch": 0.75,
2736
+ "grad_norm": 0.0039563101119373,
2737
+ "learning_rate": 0.031074166151605298,
2738
+ "loss": 1.2983,
2739
+ "step": 390
2740
+ },
2741
+ {
2742
+ "epoch": 0.7519230769230769,
2743
+ "grad_norm": 0.004276787687338926,
2744
+ "learning_rate": 0.030623894397145836,
2745
+ "loss": 1.3824,
2746
+ "step": 391
2747
+ },
2748
+ {
2749
+ "epoch": 0.7538461538461538,
2750
+ "grad_norm": 0.0038780606561958524,
2751
+ "learning_rate": 0.03017631819139273,
2752
+ "loss": 1.1824,
2753
+ "step": 392
2754
+ },
2755
+ {
2756
+ "epoch": 0.7557692307692307,
2757
+ "grad_norm": 0.003843905920764169,
2758
+ "learning_rate": 0.029731454924533086,
2759
+ "loss": 1.249,
2760
+ "step": 393
2761
+ },
2762
+ {
2763
+ "epoch": 0.7576923076923077,
2764
+ "grad_norm": 0.004438054783930797,
2765
+ "learning_rate": 0.029289321881345254,
2766
+ "loss": 1.2514,
2767
+ "step": 394
2768
+ },
2769
+ {
2770
+ "epoch": 0.7596153846153846,
2771
+ "grad_norm": 0.00414696219270615,
2772
+ "learning_rate": 0.028849936240527008,
2773
+ "loss": 1.2056,
2774
+ "step": 395
2775
+ },
2776
+ {
2777
+ "epoch": 0.7615384615384615,
2778
+ "grad_norm": 0.004198495861727321,
2779
+ "learning_rate": 0.028413315074028157,
2780
+ "loss": 1.3015,
2781
+ "step": 396
2782
+ },
2783
+ {
2784
+ "epoch": 0.7634615384615384,
2785
+ "grad_norm": 0.0037988107939604943,
2786
+ "learning_rate": 0.027979475346387363,
2787
+ "loss": 1.2832,
2788
+ "step": 397
2789
+ },
2790
+ {
2791
+ "epoch": 0.7653846153846153,
2792
+ "grad_norm": 0.004072229810402928,
2793
+ "learning_rate": 0.027548433914072735,
2794
+ "loss": 1.2722,
2795
+ "step": 398
2796
+ },
2797
+ {
2798
+ "epoch": 0.7673076923076924,
2799
+ "grad_norm": 0.004099850772340914,
2800
+ "learning_rate": 0.027120207524827168,
2801
+ "loss": 1.3084,
2802
+ "step": 399
2803
+ },
2804
+ {
2805
+ "epoch": 0.7692307692307693,
2806
+ "grad_norm": 0.00427965604011675,
2807
+ "learning_rate": 0.02669481281701739,
2808
+ "loss": 1.3743,
2809
+ "step": 400
2810
+ },
2811
+ {
2812
+ "epoch": 0.7711538461538462,
2813
+ "grad_norm": 0.003992091355850892,
2814
+ "learning_rate": 0.026272266318987603,
2815
+ "loss": 1.0924,
2816
+ "step": 401
2817
+ },
2818
+ {
2819
+ "epoch": 0.7730769230769231,
2820
+ "grad_norm": 0.004138452444748428,
2821
+ "learning_rate": 0.02585258444841733,
2822
+ "loss": 1.2157,
2823
+ "step": 402
2824
+ },
2825
+ {
2826
+ "epoch": 0.775,
2827
+ "grad_norm": 0.004524497822569178,
2828
+ "learning_rate": 0.025435783511683442,
2829
+ "loss": 1.2509,
2830
+ "step": 403
2831
+ },
2832
+ {
2833
+ "epoch": 0.7769230769230769,
2834
+ "grad_norm": 0.005042566691172367,
2835
+ "learning_rate": 0.02502187970322657,
2836
+ "loss": 1.1588,
2837
+ "step": 404
2838
+ },
2839
+ {
2840
+ "epoch": 0.7788461538461539,
2841
+ "grad_norm": 0.0038333340361418346,
2842
+ "learning_rate": 0.02461088910492202,
2843
+ "loss": 1.3019,
2844
+ "step": 405
2845
+ },
2846
+ {
2847
+ "epoch": 0.7807692307692308,
2848
+ "grad_norm": 0.004675341550863376,
2849
+ "learning_rate": 0.02420282768545469,
2850
+ "loss": 1.2414,
2851
+ "step": 406
2852
+ },
2853
+ {
2854
+ "epoch": 0.7826923076923077,
2855
+ "grad_norm": 0.00430925196868412,
2856
+ "learning_rate": 0.02379771129969892,
2857
+ "loss": 1.3508,
2858
+ "step": 407
2859
+ },
2860
+ {
2861
+ "epoch": 0.7846153846153846,
2862
+ "grad_norm": 0.004278791144378458,
2863
+ "learning_rate": 0.023395555688102213,
2864
+ "loss": 1.2416,
2865
+ "step": 408
2866
+ },
2867
+ {
2868
+ "epoch": 0.7865384615384615,
2869
+ "grad_norm": 0.004565226886170772,
2870
+ "learning_rate": 0.02299637647607372,
2871
+ "loss": 1.3721,
2872
+ "step": 409
2873
+ },
2874
+ {
2875
+ "epoch": 0.7884615384615384,
2876
+ "grad_norm": 0.004228122455231795,
2877
+ "learning_rate": 0.022600189173377264,
2878
+ "loss": 1.0756,
2879
+ "step": 410
2880
+ },
2881
+ {
2882
+ "epoch": 0.7903846153846154,
2883
+ "grad_norm": 0.004344511022418234,
2884
+ "learning_rate": 0.022207009173528525,
2885
+ "loss": 1.3352,
2886
+ "step": 411
2887
+ },
2888
+ {
2889
+ "epoch": 0.7923076923076923,
2890
+ "grad_norm": 0.0040948261335982845,
2891
+ "learning_rate": 0.02181685175319702,
2892
+ "loss": 1.2474,
2893
+ "step": 412
2894
+ },
2895
+ {
2896
+ "epoch": 0.7942307692307692,
2897
+ "grad_norm": 0.004326636290207469,
2898
+ "learning_rate": 0.021429732071612653,
2899
+ "loss": 1.342,
2900
+ "step": 413
2901
+ },
2902
+ {
2903
+ "epoch": 0.7961538461538461,
2904
+ "grad_norm": 0.0038293374498136103,
2905
+ "learning_rate": 0.02104566516997647,
2906
+ "loss": 1.1252,
2907
+ "step": 414
2908
+ },
2909
+ {
2910
+ "epoch": 0.7980769230769231,
2911
+ "grad_norm": 0.003913326125662806,
2912
+ "learning_rate": 0.020664665970876496,
2913
+ "loss": 1.2271,
2914
+ "step": 415
2915
+ },
2916
+ {
2917
+ "epoch": 0.8,
2918
+ "grad_norm": 0.004832123481233092,
2919
+ "learning_rate": 0.020286749277707784,
2920
+ "loss": 1.1393,
2921
+ "step": 416
2922
+ },
2923
+ {
2924
+ "epoch": 0.801923076923077,
2925
+ "grad_norm": 0.004833233999118917,
2926
+ "learning_rate": 0.019911929774097215,
2927
+ "loss": 1.3157,
2928
+ "step": 417
2929
+ },
2930
+ {
2931
+ "epoch": 0.8038461538461539,
2932
+ "grad_norm": 0.003775140606725514,
2933
+ "learning_rate": 0.019540222023333165,
2934
+ "loss": 1.2901,
2935
+ "step": 418
2936
+ },
2937
+ {
2938
+ "epoch": 0.8057692307692308,
2939
+ "grad_norm": 0.004370909430537529,
2940
+ "learning_rate": 0.01917164046779948,
2941
+ "loss": 1.2823,
2942
+ "step": 419
2943
+ },
2944
+ {
2945
+ "epoch": 0.8076923076923077,
2946
+ "grad_norm": 0.004029994467401848,
2947
+ "learning_rate": 0.018806199428414352,
2948
+ "loss": 1.1587,
2949
+ "step": 420
2950
+ },
2951
+ {
2952
+ "epoch": 0.8096153846153846,
2953
+ "grad_norm": 0.004181997369513217,
2954
+ "learning_rate": 0.018443913104073985,
2955
+ "loss": 1.0935,
2956
+ "step": 421
2957
+ },
2958
+ {
2959
+ "epoch": 0.8115384615384615,
2960
+ "grad_norm": 0.004329825960734355,
2961
+ "learning_rate": 0.01808479557110081,
2962
+ "loss": 1.222,
2963
+ "step": 422
2964
+ },
2965
+ {
2966
+ "epoch": 0.8134615384615385,
2967
+ "grad_norm": 0.004803369993034784,
2968
+ "learning_rate": 0.017728860782696667,
2969
+ "loss": 1.2164,
2970
+ "step": 423
2971
+ },
2972
+ {
2973
+ "epoch": 0.8153846153846154,
2974
+ "grad_norm": 0.004647999219102386,
2975
+ "learning_rate": 0.017376122568400532,
2976
+ "loss": 1.4269,
2977
+ "step": 424
2978
+ },
2979
+ {
2980
+ "epoch": 0.8173076923076923,
2981
+ "grad_norm": 0.0038943271206646965,
2982
+ "learning_rate": 0.017026594633551252,
2983
+ "loss": 1.2199,
2984
+ "step": 425
2985
+ },
2986
+ {
2987
+ "epoch": 0.8192307692307692,
2988
+ "grad_norm": 0.005554709243763775,
2989
+ "learning_rate": 0.01668029055875512,
2990
+ "loss": 1.237,
2991
+ "step": 426
2992
+ },
2993
+ {
2994
+ "epoch": 0.8211538461538461,
2995
+ "grad_norm": 0.004426076024269201,
2996
+ "learning_rate": 0.016337223799358026,
2997
+ "loss": 1.157,
2998
+ "step": 427
2999
+ },
3000
+ {
3001
+ "epoch": 0.823076923076923,
3002
+ "grad_norm": 0.004110303091180217,
3003
+ "learning_rate": 0.01599740768492286,
3004
+ "loss": 1.1239,
3005
+ "step": 428
3006
+ },
3007
+ {
3008
+ "epoch": 0.825,
3009
+ "grad_norm": 0.003802305823968438,
3010
+ "learning_rate": 0.015660855418711452,
3011
+ "loss": 1.2277,
3012
+ "step": 429
3013
+ },
3014
+ {
3015
+ "epoch": 0.8269230769230769,
3016
+ "grad_norm": 0.0037077620764046867,
3017
+ "learning_rate": 0.015327580077171589,
3018
+ "loss": 1.2256,
3019
+ "step": 430
3020
+ },
3021
+ {
3022
+ "epoch": 0.8288461538461539,
3023
+ "grad_norm": 0.004658504404892528,
3024
+ "learning_rate": 0.014997594609429088,
3025
+ "loss": 1.2969,
3026
+ "step": 431
3027
+ },
3028
+ {
3029
+ "epoch": 0.8307692307692308,
3030
+ "grad_norm": 0.004453534470209584,
3031
+ "learning_rate": 0.01467091183678444,
3032
+ "loss": 1.1286,
3033
+ "step": 432
3034
+ },
3035
+ {
3036
+ "epoch": 0.8326923076923077,
3037
+ "grad_norm": 0.00411278422557371,
3038
+ "learning_rate": 0.014347544452214867,
3039
+ "loss": 1.2702,
3040
+ "step": 433
3041
+ },
3042
+ {
3043
+ "epoch": 0.8346153846153846,
3044
+ "grad_norm": 0.0037888525040605304,
3045
+ "learning_rate": 0.014027505019880971,
3046
+ "loss": 1.008,
3047
+ "step": 434
3048
+ },
3049
+ {
3050
+ "epoch": 0.8365384615384616,
3051
+ "grad_norm": 0.0045604757237192916,
3052
+ "learning_rate": 0.013710805974638696,
3053
+ "loss": 1.3073,
3054
+ "step": 435
3055
+ },
3056
+ {
3057
+ "epoch": 0.8384615384615385,
3058
+ "grad_norm": 0.004026810861794217,
3059
+ "learning_rate": 0.01339745962155613,
3060
+ "loss": 1.0999,
3061
+ "step": 436
3062
+ },
3063
+ {
3064
+ "epoch": 0.8403846153846154,
3065
+ "grad_norm": 0.00395694256588988,
3066
+ "learning_rate": 0.01308747813543536,
3067
+ "loss": 1.3308,
3068
+ "step": 437
3069
+ },
3070
+ {
3071
+ "epoch": 0.8423076923076923,
3072
+ "grad_norm": 0.0038722492315842556,
3073
+ "learning_rate": 0.012780873560339467,
3074
+ "loss": 1.1319,
3075
+ "step": 438
3076
+ },
3077
+ {
3078
+ "epoch": 0.8442307692307692,
3079
+ "grad_norm": 0.0034959926759151744,
3080
+ "learning_rate": 0.012477657809124632,
3081
+ "loss": 1.2639,
3082
+ "step": 439
3083
+ },
3084
+ {
3085
+ "epoch": 0.8461538461538461,
3086
+ "grad_norm": 0.003933918134077107,
3087
+ "learning_rate": 0.012177842662977134,
3088
+ "loss": 1.1877,
3089
+ "step": 440
3090
+ },
3091
+ {
3092
+ "epoch": 0.8480769230769231,
3093
+ "grad_norm": 0.00966909500773923,
3094
+ "learning_rate": 0.01188143977095576,
3095
+ "loss": 1.3311,
3096
+ "step": 441
3097
+ },
3098
+ {
3099
+ "epoch": 0.85,
3100
+ "grad_norm": 0.0047254894112172185,
3101
+ "learning_rate": 0.011588460649539035,
3102
+ "loss": 1.2439,
3103
+ "step": 442
3104
+ },
3105
+ {
3106
+ "epoch": 0.8519230769230769,
3107
+ "grad_norm": 0.004369517954831794,
3108
+ "learning_rate": 0.011298916682177829,
3109
+ "loss": 1.2637,
3110
+ "step": 443
3111
+ },
3112
+ {
3113
+ "epoch": 0.8538461538461538,
3114
+ "grad_norm": 0.0036104841987051137,
3115
+ "learning_rate": 0.011012819118853146,
3116
+ "loss": 1.2283,
3117
+ "step": 444
3118
+ },
3119
+ {
3120
+ "epoch": 0.8557692307692307,
3121
+ "grad_norm": 0.004229853883737942,
3122
+ "learning_rate": 0.01073017907563887,
3123
+ "loss": 1.1472,
3124
+ "step": 445
3125
+ },
3126
+ {
3127
+ "epoch": 0.8576923076923076,
3128
+ "grad_norm": 0.003944665199706204,
3129
+ "learning_rate": 0.010451007534269908,
3130
+ "loss": 1.3755,
3131
+ "step": 446
3132
+ },
3133
+ {
3134
+ "epoch": 0.8596153846153847,
3135
+ "grad_norm": 0.004005018900200279,
3136
+ "learning_rate": 0.010175315341715598,
3137
+ "loss": 1.2436,
3138
+ "step": 447
3139
+ },
3140
+ {
3141
+ "epoch": 0.8615384615384616,
3142
+ "grad_norm": 0.004515116101782461,
3143
+ "learning_rate": 0.009903113209758098,
3144
+ "loss": 1.2136,
3145
+ "step": 448
3146
+ },
3147
+ {
3148
+ "epoch": 0.8634615384615385,
3149
+ "grad_norm": 0.0043348127453525604,
3150
+ "learning_rate": 0.009634411714576352,
3151
+ "loss": 1.3329,
3152
+ "step": 449
3153
+ },
3154
+ {
3155
+ "epoch": 0.8653846153846154,
3156
+ "grad_norm": 0.004065147932622601,
3157
+ "learning_rate": 0.009369221296335007,
3158
+ "loss": 1.2662,
3159
+ "step": 450
3160
+ },
3161
+ {
3162
+ "epoch": 0.8673076923076923,
3163
+ "grad_norm": 0.004111201868593292,
3164
+ "learning_rate": 0.009107552258778906,
3165
+ "loss": 1.2501,
3166
+ "step": 451
3167
+ },
3168
+ {
3169
+ "epoch": 0.8692307692307693,
3170
+ "grad_norm": 0.004506456425976872,
3171
+ "learning_rate": 0.008849414768832687,
3172
+ "loss": 1.3724,
3173
+ "step": 452
3174
+ },
3175
+ {
3176
+ "epoch": 0.8711538461538462,
3177
+ "grad_norm": 0.004494079470636866,
3178
+ "learning_rate": 0.008594818856205699,
3179
+ "loss": 1.3501,
3180
+ "step": 453
3181
+ },
3182
+ {
3183
+ "epoch": 0.8730769230769231,
3184
+ "grad_norm": 0.004357265321762396,
3185
+ "learning_rate": 0.00834377441300238,
3186
+ "loss": 1.1666,
3187
+ "step": 454
3188
+ },
3189
+ {
3190
+ "epoch": 0.875,
3191
+ "grad_norm": 0.0039994554491281096,
3192
+ "learning_rate": 0.008096291193337934,
3193
+ "loss": 1.2957,
3194
+ "step": 455
3195
+ },
3196
+ {
3197
+ "epoch": 0.8769230769230769,
3198
+ "grad_norm": 0.004934057471575709,
3199
+ "learning_rate": 0.007852378812959226,
3200
+ "loss": 1.2168,
3201
+ "step": 456
3202
+ },
3203
+ {
3204
+ "epoch": 0.8788461538461538,
3205
+ "grad_norm": 0.004945775175925625,
3206
+ "learning_rate": 0.007612046748871327,
3207
+ "loss": 1.33,
3208
+ "step": 457
3209
+ },
3210
+ {
3211
+ "epoch": 0.8807692307692307,
3212
+ "grad_norm": 0.004231602107070679,
3213
+ "learning_rate": 0.007375304338969136,
3214
+ "loss": 1.3634,
3215
+ "step": 458
3216
+ },
3217
+ {
3218
+ "epoch": 0.8826923076923077,
3219
+ "grad_norm": 0.0038873402701475285,
3220
+ "learning_rate": 0.007142160781674645,
3221
+ "loss": 1.2907,
3222
+ "step": 459
3223
+ },
3224
+ {
3225
+ "epoch": 0.8846153846153846,
3226
+ "grad_norm": 0.003798562909595663,
3227
+ "learning_rate": 0.006912625135579587,
3228
+ "loss": 1.1613,
3229
+ "step": 460
3230
+ },
3231
+ {
3232
+ "epoch": 0.8865384615384615,
3233
+ "grad_norm": 0.0040348980101575435,
3234
+ "learning_rate": 0.0066867063190933496,
3235
+ "loss": 1.4095,
3236
+ "step": 461
3237
+ },
3238
+ {
3239
+ "epoch": 0.8884615384615384,
3240
+ "grad_norm": 0.00393982387683152,
3241
+ "learning_rate": 0.006464413110096601,
3242
+ "loss": 1.4231,
3243
+ "step": 462
3244
+ },
3245
+ {
3246
+ "epoch": 0.8903846153846153,
3247
+ "grad_norm": 0.004172560387829518,
3248
+ "learning_rate": 0.006245754145600091,
3249
+ "loss": 1.1293,
3250
+ "step": 463
3251
+ },
3252
+ {
3253
+ "epoch": 0.8923076923076924,
3254
+ "grad_norm": 0.004097511898697587,
3255
+ "learning_rate": 0.006030737921409169,
3256
+ "loss": 1.2707,
3257
+ "step": 464
3258
+ },
3259
+ {
3260
+ "epoch": 0.8942307692307693,
3261
+ "grad_norm": 0.004544121095712231,
3262
+ "learning_rate": 0.005819372791793654,
3263
+ "loss": 1.3974,
3264
+ "step": 465
3265
+ },
3266
+ {
3267
+ "epoch": 0.8961538461538462,
3268
+ "grad_norm": 0.0035290798848328004,
3269
+ "learning_rate": 0.005611666969163243,
3270
+ "loss": 1.2574,
3271
+ "step": 466
3272
+ },
3273
+ {
3274
+ "epoch": 0.8980769230769231,
3275
+ "grad_norm": 0.004066551732880179,
3276
+ "learning_rate": 0.005407628523748398,
3277
+ "loss": 1.2974,
3278
+ "step": 467
3279
+ },
3280
+ {
3281
+ "epoch": 0.9,
3282
+ "grad_norm": 0.004456425765302806,
3283
+ "learning_rate": 0.00520726538328683,
3284
+ "loss": 1.2402,
3285
+ "step": 468
3286
+ },
3287
+ {
3288
+ "epoch": 0.9019230769230769,
3289
+ "grad_norm": 0.004182933972075675,
3290
+ "learning_rate": 0.005010585332715401,
3291
+ "loss": 1.2901,
3292
+ "step": 469
3293
+ },
3294
+ {
3295
+ "epoch": 0.9038461538461539,
3296
+ "grad_norm": 0.0034990425134677258,
3297
+ "learning_rate": 0.004817596013867765,
3298
+ "loss": 1.1613,
3299
+ "step": 470
3300
+ },
3301
+ {
3302
+ "epoch": 0.9057692307692308,
3303
+ "grad_norm": 0.003922024131287971,
3304
+ "learning_rate": 0.004628304925177318,
3305
+ "loss": 1.1798,
3306
+ "step": 471
3307
+ },
3308
+ {
3309
+ "epoch": 0.9076923076923077,
3310
+ "grad_norm": 0.003920402470152831,
3311
+ "learning_rate": 0.004442719421385921,
3312
+ "loss": 1.1588,
3313
+ "step": 472
3314
+ },
3315
+ {
3316
+ "epoch": 0.9096153846153846,
3317
+ "grad_norm": 0.004079270415627159,
3318
+ "learning_rate": 0.004260846713258193,
3319
+ "loss": 1.2123,
3320
+ "step": 473
3321
+ },
3322
+ {
3323
+ "epoch": 0.9115384615384615,
3324
+ "grad_norm": 0.0037684877296426727,
3325
+ "learning_rate": 0.004082693867301224,
3326
+ "loss": 1.3289,
3327
+ "step": 474
3328
+ },
3329
+ {
3330
+ "epoch": 0.9134615384615384,
3331
+ "grad_norm": 0.0038378678547793437,
3332
+ "learning_rate": 0.003908267805490051,
3333
+ "loss": 1.2432,
3334
+ "step": 475
3335
+ },
3336
+ {
3337
+ "epoch": 0.9153846153846154,
3338
+ "grad_norm": 0.004418071495125169,
3339
+ "learning_rate": 0.003737575304998797,
3340
+ "loss": 1.2127,
3341
+ "step": 476
3342
+ },
3343
+ {
3344
+ "epoch": 0.9173076923076923,
3345
+ "grad_norm": 0.004663534673924085,
3346
+ "learning_rate": 0.003570622997937234,
3347
+ "loss": 1.2001,
3348
+ "step": 477
3349
+ },
3350
+ {
3351
+ "epoch": 0.9192307692307692,
3352
+ "grad_norm": 0.003947802387460568,
3353
+ "learning_rate": 0.00340741737109318,
3354
+ "loss": 1.1586,
3355
+ "step": 478
3356
+ },
3357
+ {
3358
+ "epoch": 0.9211538461538461,
3359
+ "grad_norm": 0.004217210255715102,
3360
+ "learning_rate": 0.003247964765680389,
3361
+ "loss": 1.3071,
3362
+ "step": 479
3363
+ },
3364
+ {
3365
+ "epoch": 0.9230769230769231,
3366
+ "grad_norm": 0.004283612997178901,
3367
+ "learning_rate": 0.0030922713770922153,
3368
+ "loss": 1.321,
3369
+ "step": 480
3370
+ },
3371
+ {
3372
+ "epoch": 0.925,
3373
+ "grad_norm": 0.0039971533834376495,
3374
+ "learning_rate": 0.0029403432546609046,
3375
+ "loss": 1.3411,
3376
+ "step": 481
3377
+ },
3378
+ {
3379
+ "epoch": 0.926923076923077,
3380
+ "grad_norm": 0.004460163304059353,
3381
+ "learning_rate": 0.0027921863014225504,
3382
+ "loss": 1.3445,
3383
+ "step": 482
3384
+ },
3385
+ {
3386
+ "epoch": 0.9288461538461539,
3387
+ "grad_norm": 0.004473241983110544,
3388
+ "learning_rate": 0.002647806273887665,
3389
+ "loss": 1.2325,
3390
+ "step": 483
3391
+ },
3392
+ {
3393
+ "epoch": 0.9307692307692308,
3394
+ "grad_norm": 0.003940002054689835,
3395
+ "learning_rate": 0.0025072087818176383,
3396
+ "loss": 1.2365,
3397
+ "step": 484
3398
+ },
3399
+ {
3400
+ "epoch": 0.9326923076923077,
3401
+ "grad_norm": 0.003904886029255359,
3402
+ "learning_rate": 0.002370399288006664,
3403
+ "loss": 1.1777,
3404
+ "step": 485
3405
+ },
3406
+ {
3407
+ "epoch": 0.9346153846153846,
3408
+ "grad_norm": 0.004342867799105381,
3409
+ "learning_rate": 0.0022373831080695463,
3410
+ "loss": 1.3129,
3411
+ "step": 486
3412
+ },
3413
+ {
3414
+ "epoch": 0.9365384615384615,
3415
+ "grad_norm": 0.0038287335490306525,
3416
+ "learning_rate": 0.0021081654102351635,
3417
+ "loss": 1.1561,
3418
+ "step": 487
3419
+ },
3420
+ {
3421
+ "epoch": 0.9384615384615385,
3422
+ "grad_norm": 0.00413001481486949,
3423
+ "learning_rate": 0.0019827512151456175,
3424
+ "loss": 1.0972,
3425
+ "step": 488
3426
+ },
3427
+ {
3428
+ "epoch": 0.9403846153846154,
3429
+ "grad_norm": 0.0034586554468023552,
3430
+ "learning_rate": 0.0018611453956612345,
3431
+ "loss": 1.3241,
3432
+ "step": 489
3433
+ },
3434
+ {
3435
+ "epoch": 0.9423076923076923,
3436
+ "grad_norm": 0.0041273900814665886,
3437
+ "learning_rate": 0.0017433526766711727,
3438
+ "loss": 1.2226,
3439
+ "step": 490
3440
+ },
3441
+ {
3442
+ "epoch": 0.9442307692307692,
3443
+ "grad_norm": 0.004110509837165172,
3444
+ "learning_rate": 0.0016293776349098677,
3445
+ "loss": 1.185,
3446
+ "step": 491
3447
+ },
3448
+ {
3449
+ "epoch": 0.9461538461538461,
3450
+ "grad_norm": 0.004244277277612668,
3451
+ "learning_rate": 0.0015192246987791981,
3452
+ "loss": 1.3045,
3453
+ "step": 492
3454
+ },
3455
+ {
3456
+ "epoch": 0.948076923076923,
3457
+ "grad_norm": 0.00438125624271381,
3458
+ "learning_rate": 0.0014128981481764114,
3459
+ "loss": 1.3831,
3460
+ "step": 493
3461
+ },
3462
+ {
3463
+ "epoch": 0.95,
3464
+ "grad_norm": 0.003702069552472066,
3465
+ "learning_rate": 0.0013104021143278911,
3466
+ "loss": 1.2447,
3467
+ "step": 494
3468
+ },
3469
+ {
3470
+ "epoch": 0.9519230769230769,
3471
+ "grad_norm": 0.003931542540593198,
3472
+ "learning_rate": 0.0012117405796285285,
3473
+ "loss": 1.1951,
3474
+ "step": 495
3475
+ },
3476
+ {
3477
+ "epoch": 0.9538461538461539,
3478
+ "grad_norm": 0.004208056721933632,
3479
+ "learning_rate": 0.0011169173774871477,
3480
+ "loss": 1.1145,
3481
+ "step": 496
3482
+ },
3483
+ {
3484
+ "epoch": 0.9557692307692308,
3485
+ "grad_norm": 0.003578958448552289,
3486
+ "learning_rate": 0.0010259361921774012,
3487
+ "loss": 1.2578,
3488
+ "step": 497
3489
+ },
3490
+ {
3491
+ "epoch": 0.9576923076923077,
3492
+ "grad_norm": 0.004297030225667394,
3493
+ "learning_rate": 0.000938800558694719,
3494
+ "loss": 1.2044,
3495
+ "step": 498
3496
+ },
3497
+ {
3498
+ "epoch": 0.9596153846153846,
3499
+ "grad_norm": 0.0040579333095660105,
3500
+ "learning_rate": 0.0008555138626189618,
3501
+ "loss": 1.3979,
3502
+ "step": 499
3503
+ },
3504
+ {
3505
+ "epoch": 0.9615384615384616,
3506
+ "grad_norm": 0.004780125055361835,
3507
+ "learning_rate": 0.0007760793399827937,
3508
+ "loss": 1.3175,
3509
+ "step": 500
3510
+ },
3511
+ {
3512
+ "epoch": 0.9634615384615385,
3513
+ "grad_norm": 0.005579382613523984,
3514
+ "learning_rate": 0.000700500077146038,
3515
+ "loss": 1.3461,
3516
+ "step": 501
3517
+ },
3518
+ {
3519
+ "epoch": 0.9653846153846154,
3520
+ "grad_norm": 0.0037606291216108496,
3521
+ "learning_rate": 0.0006287790106757397,
3522
+ "loss": 1.2376,
3523
+ "step": 502
3524
+ },
3525
+ {
3526
+ "epoch": 0.9673076923076923,
3527
+ "grad_norm": 0.004034653062783717,
3528
+ "learning_rate": 0.0005609189272320237,
3529
+ "loss": 1.2841,
3530
+ "step": 503
3531
+ },
3532
+ {
3533
+ "epoch": 0.9692307692307692,
3534
+ "grad_norm": 0.004427045492825959,
3535
+ "learning_rate": 0.000496922463459859,
3536
+ "loss": 1.236,
3537
+ "step": 504
3538
+ },
3539
+ {
3540
+ "epoch": 0.9711538461538461,
3541
+ "grad_norm": 0.0038481198912412935,
3542
+ "learning_rate": 0.0004367921058866187,
3543
+ "loss": 1.2791,
3544
+ "step": 505
3545
+ },
3546
+ {
3547
+ "epoch": 0.9730769230769231,
3548
+ "grad_norm": 0.004406473460942015,
3549
+ "learning_rate": 0.0003805301908254455,
3550
+ "loss": 1.1885,
3551
+ "step": 506
3552
+ },
3553
+ {
3554
+ "epoch": 0.975,
3555
+ "grad_norm": 0.00370521784158096,
3556
+ "learning_rate": 0.0003281389042844918,
3557
+ "loss": 1.444,
3558
+ "step": 507
3559
+ },
3560
+ {
3561
+ "epoch": 0.9769230769230769,
3562
+ "grad_norm": 0.0038819886365180447,
3563
+ "learning_rate": 0.00027962028188198705,
3564
+ "loss": 1.3055,
3565
+ "step": 508
3566
+ },
3567
+ {
3568
+ "epoch": 0.9788461538461538,
3569
+ "grad_norm": 0.0038555801157193824,
3570
+ "learning_rate": 0.00023497620876711256,
3571
+ "loss": 1.2756,
3572
+ "step": 509
3573
+ },
3574
+ {
3575
+ "epoch": 0.9807692307692307,
3576
+ "grad_norm": 0.003916380144825126,
3577
+ "learning_rate": 0.00019420841954681523,
3578
+ "loss": 1.237,
3579
+ "step": 510
3580
+ },
3581
+ {
3582
+ "epoch": 0.9826923076923076,
3583
+ "grad_norm": 0.0037417461996873016,
3584
+ "learning_rate": 0.00015731849821833956,
3585
+ "loss": 1.1995,
3586
+ "step": 511
3587
+ },
3588
+ {
3589
+ "epoch": 0.9846153846153847,
3590
+ "grad_norm": 0.0038150455506307476,
3591
+ "learning_rate": 0.00012430787810776555,
3592
+ "loss": 1.0845,
3593
+ "step": 512
3594
+ },
3595
+ {
3596
+ "epoch": 0.9865384615384616,
3597
+ "grad_norm": 0.004291702428498065,
3598
+ "learning_rate": 9.517784181422018e-05,
3599
+ "loss": 1.2972,
3600
+ "step": 513
3601
+ },
3602
+ {
3603
+ "epoch": 0.9884615384615385,
3604
+ "grad_norm": 0.0036050787242626926,
3605
+ "learning_rate": 6.992952116013917e-05,
3606
+ "loss": 1.2666,
3607
+ "step": 514
3608
+ },
3609
+ {
3610
+ "epoch": 0.9903846153846154,
3611
+ "grad_norm": 0.004667016975419852,
3612
+ "learning_rate": 4.856389714723575e-05,
3613
+ "loss": 1.3222,
3614
+ "step": 515
3615
+ },
3616
+ {
3617
+ "epoch": 0.9923076923076923,
3618
+ "grad_norm": 0.0038489945122420583,
3619
+ "learning_rate": 3.108179991837545e-05,
3620
+ "loss": 1.1982,
3621
+ "step": 516
3622
+ },
3623
+ {
3624
+ "epoch": 0.9942307692307693,
3625
+ "grad_norm": 0.004250207716393607,
3626
+ "learning_rate": 1.7483908725357544e-05,
3627
+ "loss": 1.364,
3628
+ "step": 517
3629
+ },
3630
+ {
3631
+ "epoch": 0.9961538461538462,
3632
+ "grad_norm": 0.004090889724010192,
3633
+ "learning_rate": 7.770751902513862e-06,
3634
+ "loss": 1.2226,
3635
+ "step": 518
3636
+ },
3637
+ {
3638
+ "epoch": 0.9980769230769231,
3639
+ "grad_norm": 0.003902358225511784,
3640
+ "learning_rate": 1.9427068461808086e-06,
3641
+ "loss": 1.2963,
3642
+ "step": 519
3643
+ },
3644
+ {
3645
+ "epoch": 1.0,
3646
+ "grad_norm": 0.004540588422952889,
3647
+ "learning_rate": 0.0,
3648
+ "loss": 1.3767,
3649
+ "step": 520
3650
+ },
3651
+ {
3652
+ "epoch": 1.0,
3653
+ "step": 520,
3654
+ "total_flos": 419724754747392.0,
3655
+ "train_loss": 1.5168263238209945,
3656
+ "train_runtime": 1931.5,
3657
+ "train_samples_per_second": 34.444,
3658
+ "train_steps_per_second": 0.269
3659
+ }
3660
+ ],
3661
+ "logging_steps": 1.0,
3662
+ "max_steps": 520,
3663
+ "num_input_tokens_seen": 0,
3664
+ "num_train_epochs": 1,
3665
+ "save_steps": 50000,
3666
+ "total_flos": 419724754747392.0,
3667
+ "train_batch_size": 4,
3668
+ "trial_name": null,
3669
+ "trial_params": null
3670
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:505ad3e7682e4228a417f4b17f4ded295491451ddb1220c7b15869ab8a0bccd2
3
+ size 7608
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_1.7_2e-1_connector-1.0_1.7_2e-1_ablation/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TinyLlavaForConditionalGeneration"
4
+ ],
5
+ "backward_type_connector": "normal",
6
+ "cache_dir": null,
7
+ "connector_type": "mlp2x_gelu",
8
+ "hidden_size": 896,
9
+ "ignore_index": -100,
10
+ "image_aspect_ratio": "square",
11
+ "image_token_index": -200,
12
+ "llm_model_name_or_path": "Qwen/Qwen2.5-0.5B",
13
+ "mask_model": [
14
+ "llm",
15
+ "connector"
16
+ ],
17
+ "mask_type_connector": "soft",
18
+ "model_type": "tinyllava",
19
+ "num_queries": 128,
20
+ "num_resampler_layers": 3,
21
+ "pad_token": "<|endoftext|>",
22
+ "pad_token_id": 151643,
23
+ "resampler_hidden_size": 768,
24
+ "sparsity_connector": null,
25
+ "subnet_type_connector": "global",
26
+ "temperature_connector": 2.3,
27
+ "text_config": {
28
+ "_name_or_path": "Qwen/Qwen2.5-0.5B",
29
+ "architectures": [
30
+ "Qwen2ForCausalLM"
31
+ ],
32
+ "backward_type": "normal",
33
+ "bos_token_id": 151643,
34
+ "eos_token_id": 151643,
35
+ "hidden_size": 896,
36
+ "intermediate_size": 4864,
37
+ "mask_type": "soft",
38
+ "masked_layers": "all",
39
+ "max_position_embeddings": 32768,
40
+ "max_window_layers": 24,
41
+ "model_type": "qwen2",
42
+ "num_attention_heads": 14,
43
+ "num_hidden_layers": 24,
44
+ "num_key_value_heads": 2,
45
+ "rope_theta": 1000000.0,
46
+ "sliding_window": 32768,
47
+ "subnet_mode": "both",
48
+ "subnet_type": "None",
49
+ "temperature_attn": 2.3,
50
+ "temperature_mlp": 2.3,
51
+ "tie_word_embeddings": true,
52
+ "torch_dtype": "bfloat16",
53
+ "use_mrope": false,
54
+ "use_sliding_window": false,
55
+ "vocab_size": 151936
56
+ },
57
+ "threshold_connector": null,
58
+ "tokenizer_model_max_length": 2048,
59
+ "tokenizer_name_or_path": "Qwen/Qwen2.5-0.5B",
60
+ "tokenizer_padding_side": "right",
61
+ "tokenizer_use_fast": false,
62
+ "torch_dtype": "bfloat16",
63
+ "transformers_version": "4.40.1",
64
+ "tune_type_connector": "full",
65
+ "tune_type_llm": "full",
66
+ "tune_type_vision_tower": "frozen",
67
+ "tune_vision_tower_from_layer": 0,
68
+ "use_cache": true,
69
+ "vision_config": {
70
+ "hidden_act": "gelu_pytorch_tanh",
71
+ "hidden_size": 1152,
72
+ "image_size": 384,
73
+ "intermediate_size": 4304,
74
+ "layer_norm_eps": 1e-06,
75
+ "model_name_or_path": "google/siglip-so400m-patch14-384",
76
+ "model_name_or_path2": "",
77
+ "model_type": "siglip_vision_model",
78
+ "num_attention_heads": 16,
79
+ "num_hidden_layers": 27,
80
+ "patch_size": 14
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "patch",
84
+ "vision_hidden_size": 1152,
85
+ "vision_model_name_or_path": "google/siglip-so400m-patch14-384",
86
+ "vision_model_name_or_path2": "",
87
+ "vocab_size": 151936
88
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151643,
4
+ "eos_token_id": 151643,
5
+ "transformers_version": "4.40.1",
6
+ "use_cache": false
7
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/log.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ 2025-10-17 08:52:17,925 | INFO: Total Parameters: 1283756736, Total Trainable Parameters: 359661568
2
+ 2025-10-17 08:52:17,931 | INFO: Trainable Parameters:
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20d3e9a8b0c4bd2fb2abaffca6bdcbbc45a46bfb9d0aec2102a896e679d3d36c
3
+ size 2567640568
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/runs/Oct17_08-51-11_ywang29-vrdb-test1-worker-0/events.out.tfevents.1760691142.ywang29-vrdb-test1-worker-0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c9c05031e9ab87891f62d38d53055398f23f941198fd881f884c88c0d0dfece
3
+ size 116893
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/special_tokens_map.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ },
31
+ "unk_token": "<|endoftext|>"
32
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/tokenizer_config.json ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|endoftext|>",
201
+ "errors": "replace",
202
+ "model_max_length": 2048,
203
+ "pad_token": "<|endoftext|>",
204
+ "padding_side": "right",
205
+ "split_special_tokens": false,
206
+ "tokenizer_class": "Qwen2Tokenizer",
207
+ "unk_token": "<|endoftext|>"
208
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/trainer_state.json ADDED
@@ -0,0 +1,3670 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 520,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0019230769230769232,
13
+ "grad_norm": 0.39135470954371554,
14
+ "learning_rate": 0.0125,
15
+ "loss": 8.5235,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0038461538461538464,
20
+ "grad_norm": 0.40033961788993977,
21
+ "learning_rate": 0.025,
22
+ "loss": 7.7698,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.0057692307692307696,
27
+ "grad_norm": 0.20009993755824826,
28
+ "learning_rate": 0.037500000000000006,
29
+ "loss": 7.1513,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.007692307692307693,
34
+ "grad_norm": 0.11811626816804653,
35
+ "learning_rate": 0.05,
36
+ "loss": 6.3043,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.009615384615384616,
41
+ "grad_norm": 0.10193982724163131,
42
+ "learning_rate": 0.0625,
43
+ "loss": 5.2741,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.011538461538461539,
48
+ "grad_norm": 0.4594015829521082,
49
+ "learning_rate": 0.07500000000000001,
50
+ "loss": 7.6926,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.013461538461538462,
55
+ "grad_norm": 0.06104242293848131,
56
+ "learning_rate": 0.08750000000000001,
57
+ "loss": 4.8635,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.015384615384615385,
62
+ "grad_norm": 0.06982561503938566,
63
+ "learning_rate": 0.1,
64
+ "loss": 4.6665,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.01730769230769231,
69
+ "grad_norm": 0.048657264205471756,
70
+ "learning_rate": 0.1125,
71
+ "loss": 4.1599,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.019230769230769232,
76
+ "grad_norm": 0.027900928844387687,
77
+ "learning_rate": 0.125,
78
+ "loss": 3.0313,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.021153846153846155,
83
+ "grad_norm": 0.051905999170411735,
84
+ "learning_rate": 0.1375,
85
+ "loss": 3.2826,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.023076923076923078,
90
+ "grad_norm": 0.10385900366893912,
91
+ "learning_rate": 0.15000000000000002,
92
+ "loss": 3.6371,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.025,
97
+ "grad_norm": 0.02161264412246251,
98
+ "learning_rate": 0.1625,
99
+ "loss": 2.6605,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.026923076923076925,
104
+ "grad_norm": 0.017157420040105714,
105
+ "learning_rate": 0.17500000000000002,
106
+ "loss": 2.5932,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.028846153846153848,
111
+ "grad_norm": 0.04183922224686035,
112
+ "learning_rate": 0.1875,
113
+ "loss": 2.8849,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.03076923076923077,
118
+ "grad_norm": 0.018441286029950446,
119
+ "learning_rate": 0.2,
120
+ "loss": 2.5494,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.032692307692307694,
125
+ "grad_norm": 0.018556598984477094,
126
+ "learning_rate": 0.1999980572931538,
127
+ "loss": 2.3449,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.03461538461538462,
132
+ "grad_norm": 0.019665728080086584,
133
+ "learning_rate": 0.19999222924809748,
134
+ "loss": 2.0463,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.03653846153846154,
139
+ "grad_norm": 0.08489095974042293,
140
+ "learning_rate": 0.19998251609127465,
141
+ "loss": 2.8307,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.038461538461538464,
146
+ "grad_norm": 0.04546095215678687,
147
+ "learning_rate": 0.19996891820008164,
148
+ "loss": 2.2875,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.04038461538461539,
153
+ "grad_norm": 0.037741591174263234,
154
+ "learning_rate": 0.19995143610285276,
155
+ "loss": 2.7032,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.04230769230769231,
160
+ "grad_norm": 0.01567133578190371,
161
+ "learning_rate": 0.19993007047883987,
162
+ "loss": 2.1187,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.04423076923076923,
167
+ "grad_norm": 0.012111227868178837,
168
+ "learning_rate": 0.1999048221581858,
169
+ "loss": 1.9981,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.046153846153846156,
174
+ "grad_norm": 0.017659587853928346,
175
+ "learning_rate": 0.19987569212189224,
176
+ "loss": 2.3199,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.04807692307692308,
181
+ "grad_norm": 0.020222948995125663,
182
+ "learning_rate": 0.19984268150178167,
183
+ "loss": 2.0452,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.05,
188
+ "grad_norm": 0.010819899426843842,
189
+ "learning_rate": 0.1998057915804532,
190
+ "loss": 1.9767,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.051923076923076926,
195
+ "grad_norm": 0.016860769074019715,
196
+ "learning_rate": 0.1997650237912329,
197
+ "loss": 1.8151,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.05384615384615385,
202
+ "grad_norm": 0.010766926154834378,
203
+ "learning_rate": 0.199720379718118,
204
+ "loss": 1.7633,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.05576923076923077,
209
+ "grad_norm": 0.01005679136907991,
210
+ "learning_rate": 0.19967186109571552,
211
+ "loss": 1.7409,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.057692307692307696,
216
+ "grad_norm": 0.009274872212699023,
217
+ "learning_rate": 0.19961946980917455,
218
+ "loss": 2.2413,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.05961538461538462,
223
+ "grad_norm": 0.011917804198775094,
224
+ "learning_rate": 0.1995632078941134,
225
+ "loss": 1.7404,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.06153846153846154,
230
+ "grad_norm": 0.013530083570140324,
231
+ "learning_rate": 0.19950307753654017,
232
+ "loss": 2.3335,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.06346153846153846,
237
+ "grad_norm": 0.008344376213960432,
238
+ "learning_rate": 0.19943908107276798,
239
+ "loss": 1.7131,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.06538461538461539,
244
+ "grad_norm": 0.006346593253830383,
245
+ "learning_rate": 0.19937122098932428,
246
+ "loss": 1.681,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.0673076923076923,
251
+ "grad_norm": 0.0076008903368543716,
252
+ "learning_rate": 0.19929949992285395,
253
+ "loss": 1.6675,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.06923076923076923,
258
+ "grad_norm": 0.00623800893236134,
259
+ "learning_rate": 0.19922392066001723,
260
+ "loss": 1.7925,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.07115384615384615,
265
+ "grad_norm": 0.021696657826696674,
266
+ "learning_rate": 0.19914448613738106,
267
+ "loss": 2.0516,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.07307692307692308,
272
+ "grad_norm": 0.015214291150241442,
273
+ "learning_rate": 0.1990611994413053,
274
+ "loss": 1.8782,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.075,
279
+ "grad_norm": 0.006443257285149075,
280
+ "learning_rate": 0.19897406380782262,
281
+ "loss": 1.6739,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.07692307692307693,
286
+ "grad_norm": 0.01245301532269045,
287
+ "learning_rate": 0.19888308262251286,
288
+ "loss": 1.6897,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.07884615384615384,
293
+ "grad_norm": 0.00579666606058442,
294
+ "learning_rate": 0.19878825942037148,
295
+ "loss": 1.6541,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.08076923076923077,
300
+ "grad_norm": 0.01241107232846799,
301
+ "learning_rate": 0.19868959788567211,
302
+ "loss": 1.6978,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.08269230769230769,
307
+ "grad_norm": 0.01407053084220235,
308
+ "learning_rate": 0.1985871018518236,
309
+ "loss": 1.8482,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.08461538461538462,
314
+ "grad_norm": 0.00853941645916838,
315
+ "learning_rate": 0.19848077530122082,
316
+ "loss": 1.9225,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.08653846153846154,
321
+ "grad_norm": 0.010273014041048929,
322
+ "learning_rate": 0.19837062236509015,
323
+ "loss": 1.6753,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.08846153846153847,
328
+ "grad_norm": 0.008455155376902447,
329
+ "learning_rate": 0.19825664732332884,
330
+ "loss": 1.9969,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.09038461538461538,
335
+ "grad_norm": 0.008343461759744151,
336
+ "learning_rate": 0.19813885460433878,
337
+ "loss": 1.6361,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.09230769230769231,
342
+ "grad_norm": 0.0059788370737448564,
343
+ "learning_rate": 0.19801724878485438,
344
+ "loss": 1.6214,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.09423076923076923,
349
+ "grad_norm": 0.005147724410503381,
350
+ "learning_rate": 0.19789183458976486,
351
+ "loss": 1.6173,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.09615384615384616,
356
+ "grad_norm": 0.005810256898772154,
357
+ "learning_rate": 0.19776261689193048,
358
+ "loss": 1.5947,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.09807692307692308,
363
+ "grad_norm": 0.005305437434279318,
364
+ "learning_rate": 0.19762960071199334,
365
+ "loss": 1.5187,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.1,
370
+ "grad_norm": 0.008860270639977531,
371
+ "learning_rate": 0.19749279121818236,
372
+ "loss": 1.6726,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.10192307692307692,
377
+ "grad_norm": 0.005252434257427864,
378
+ "learning_rate": 0.19735219372611235,
379
+ "loss": 1.6645,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.10384615384615385,
384
+ "grad_norm": 0.005912959391551706,
385
+ "learning_rate": 0.19720781369857746,
386
+ "loss": 1.5291,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.10576923076923077,
391
+ "grad_norm": 0.006320634200577182,
392
+ "learning_rate": 0.1970596567453391,
393
+ "loss": 1.5297,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.1076923076923077,
398
+ "grad_norm": 0.0054246690643924645,
399
+ "learning_rate": 0.1969077286229078,
400
+ "loss": 1.6759,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.10961538461538461,
405
+ "grad_norm": 0.007310240107849589,
406
+ "learning_rate": 0.19675203523431964,
407
+ "loss": 1.5059,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.11153846153846154,
412
+ "grad_norm": 0.005787522037825068,
413
+ "learning_rate": 0.19659258262890683,
414
+ "loss": 1.6837,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.11346153846153846,
419
+ "grad_norm": 0.006077949402765502,
420
+ "learning_rate": 0.19642937700206278,
421
+ "loss": 1.66,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.11538461538461539,
426
+ "grad_norm": 0.008404646974010941,
427
+ "learning_rate": 0.19626242469500121,
428
+ "loss": 1.5971,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.11730769230769231,
433
+ "grad_norm": 0.00763822006408089,
434
+ "learning_rate": 0.19609173219450998,
435
+ "loss": 1.809,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.11923076923076924,
440
+ "grad_norm": 0.0058158235170261375,
441
+ "learning_rate": 0.19591730613269878,
442
+ "loss": 1.5566,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.12115384615384615,
447
+ "grad_norm": 0.006486297503373095,
448
+ "learning_rate": 0.19573915328674182,
449
+ "loss": 1.5613,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.12307692307692308,
454
+ "grad_norm": 0.005711191418872784,
455
+ "learning_rate": 0.1955572805786141,
456
+ "loss": 1.5806,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.125,
461
+ "grad_norm": 0.005927057479767287,
462
+ "learning_rate": 0.1953716950748227,
463
+ "loss": 1.5805,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.12692307692307692,
468
+ "grad_norm": 0.006790735828584353,
469
+ "learning_rate": 0.19518240398613226,
470
+ "loss": 1.5418,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.12884615384615383,
475
+ "grad_norm": 0.005876535969410615,
476
+ "learning_rate": 0.1949894146672846,
477
+ "loss": 1.42,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.13076923076923078,
482
+ "grad_norm": 0.004667064259622374,
483
+ "learning_rate": 0.1947927346167132,
484
+ "loss": 1.4618,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.1326923076923077,
489
+ "grad_norm": 0.007846722693776387,
490
+ "learning_rate": 0.1945923714762516,
491
+ "loss": 1.4525,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.1346153846153846,
496
+ "grad_norm": 0.006204965689361106,
497
+ "learning_rate": 0.19438833303083677,
498
+ "loss": 1.5135,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.13653846153846153,
503
+ "grad_norm": 0.004466470431107101,
504
+ "learning_rate": 0.19418062720820636,
505
+ "loss": 1.4155,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.13846153846153847,
510
+ "grad_norm": 0.0058966542973343935,
511
+ "learning_rate": 0.19396926207859086,
512
+ "loss": 1.5781,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.14038461538461539,
517
+ "grad_norm": 0.004841869453844232,
518
+ "learning_rate": 0.19375424585439993,
519
+ "loss": 1.404,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.1423076923076923,
524
+ "grad_norm": 0.005291567628655341,
525
+ "learning_rate": 0.1935355868899034,
526
+ "loss": 1.5295,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.14423076923076922,
531
+ "grad_norm": 0.004661908712860183,
532
+ "learning_rate": 0.19331329368090666,
533
+ "loss": 1.4128,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.14615384615384616,
538
+ "grad_norm": 0.008944170932220627,
539
+ "learning_rate": 0.19308737486442043,
540
+ "loss": 1.8292,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.14807692307692308,
545
+ "grad_norm": 0.0055636931072980235,
546
+ "learning_rate": 0.19285783921832536,
547
+ "loss": 1.3508,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.15,
552
+ "grad_norm": 0.005336373744361852,
553
+ "learning_rate": 0.19262469566103088,
554
+ "loss": 1.4812,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.1519230769230769,
559
+ "grad_norm": 0.00463919314187253,
560
+ "learning_rate": 0.19238795325112867,
561
+ "loss": 1.4509,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.15384615384615385,
566
+ "grad_norm": 0.006331565118846022,
567
+ "learning_rate": 0.19214762118704076,
568
+ "loss": 1.7815,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.15576923076923077,
573
+ "grad_norm": 0.0059669634315569165,
574
+ "learning_rate": 0.19190370880666208,
575
+ "loss": 1.6226,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.1576923076923077,
580
+ "grad_norm": 0.004811228056077447,
581
+ "learning_rate": 0.19165622558699763,
582
+ "loss": 1.511,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.1596153846153846,
587
+ "grad_norm": 0.005317208874162749,
588
+ "learning_rate": 0.19140518114379435,
589
+ "loss": 1.5716,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.16153846153846155,
594
+ "grad_norm": 0.006793276753294022,
595
+ "learning_rate": 0.19115058523116735,
596
+ "loss": 1.561,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.16346153846153846,
601
+ "grad_norm": 0.004572375638449357,
602
+ "learning_rate": 0.1908924477412211,
603
+ "loss": 1.5667,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.16538461538461538,
608
+ "grad_norm": 0.005300092579939401,
609
+ "learning_rate": 0.19063077870366502,
610
+ "loss": 1.6028,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.1673076923076923,
615
+ "grad_norm": 0.008839660036365379,
616
+ "learning_rate": 0.1903655882854237,
617
+ "loss": 1.7407,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.16923076923076924,
622
+ "grad_norm": 0.009319479599174467,
623
+ "learning_rate": 0.19009688679024192,
624
+ "loss": 1.7601,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.17115384615384616,
629
+ "grad_norm": 0.005140909860309564,
630
+ "learning_rate": 0.18982468465828442,
631
+ "loss": 1.5296,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.17307692307692307,
636
+ "grad_norm": 0.004822523385874076,
637
+ "learning_rate": 0.1895489924657301,
638
+ "loss": 1.4592,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.175,
643
+ "grad_norm": 0.004572327116196783,
644
+ "learning_rate": 0.18926982092436118,
645
+ "loss": 1.5481,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.17692307692307693,
650
+ "grad_norm": 0.004484736101898504,
651
+ "learning_rate": 0.18898718088114688,
652
+ "loss": 1.4794,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.17884615384615385,
657
+ "grad_norm": 0.005410228455653506,
658
+ "learning_rate": 0.18870108331782218,
659
+ "loss": 1.4784,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.18076923076923077,
664
+ "grad_norm": 0.005126822034077061,
665
+ "learning_rate": 0.18841153935046098,
666
+ "loss": 1.5819,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.18269230769230768,
671
+ "grad_norm": 0.0056894339918796225,
672
+ "learning_rate": 0.18811856022904425,
673
+ "loss": 1.4589,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.18461538461538463,
678
+ "grad_norm": 0.004261508935282256,
679
+ "learning_rate": 0.18782215733702287,
680
+ "loss": 1.46,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.18653846153846154,
685
+ "grad_norm": 0.0059504206610644685,
686
+ "learning_rate": 0.18752234219087538,
687
+ "loss": 1.4294,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.18846153846153846,
692
+ "grad_norm": 0.003979760963054454,
693
+ "learning_rate": 0.18721912643966054,
694
+ "loss": 1.4264,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.19038461538461537,
699
+ "grad_norm": 0.0048284862822650905,
700
+ "learning_rate": 0.18691252186456464,
701
+ "loss": 1.4549,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.19230769230769232,
706
+ "grad_norm": 0.00738413485368456,
707
+ "learning_rate": 0.1866025403784439,
708
+ "loss": 1.6028,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.19423076923076923,
713
+ "grad_norm": 0.004692635827765809,
714
+ "learning_rate": 0.18628919402536132,
715
+ "loss": 1.4408,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.19615384615384615,
720
+ "grad_norm": 0.005424857601543698,
721
+ "learning_rate": 0.18597249498011903,
722
+ "loss": 1.4611,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.19807692307692307,
727
+ "grad_norm": 0.003991079236682591,
728
+ "learning_rate": 0.18565245554778517,
729
+ "loss": 1.3761,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.2,
734
+ "grad_norm": 0.004674133471798638,
735
+ "learning_rate": 0.18532908816321558,
736
+ "loss": 1.4636,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.20192307692307693,
741
+ "grad_norm": 0.004427948573446717,
742
+ "learning_rate": 0.18500240539057092,
743
+ "loss": 1.4623,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.20384615384615384,
748
+ "grad_norm": 0.0049963203117629265,
749
+ "learning_rate": 0.18467241992282843,
750
+ "loss": 1.588,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.20576923076923076,
755
+ "grad_norm": 0.004730500631491319,
756
+ "learning_rate": 0.18433914458128858,
757
+ "loss": 1.566,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.2076923076923077,
762
+ "grad_norm": 0.004949277456841358,
763
+ "learning_rate": 0.18400259231507718,
764
+ "loss": 1.4198,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.20961538461538462,
769
+ "grad_norm": 0.0040155664811015205,
770
+ "learning_rate": 0.18366277620064198,
771
+ "loss": 1.5396,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.21153846153846154,
776
+ "grad_norm": 0.005006118105410723,
777
+ "learning_rate": 0.1833197094412449,
778
+ "loss": 1.5959,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.21346153846153845,
783
+ "grad_norm": 0.0050266736697475636,
784
+ "learning_rate": 0.18297340536644877,
785
+ "loss": 1.6097,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.2153846153846154,
790
+ "grad_norm": 0.004274730309747852,
791
+ "learning_rate": 0.1826238774315995,
792
+ "loss": 1.4816,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.2173076923076923,
797
+ "grad_norm": 0.0038876066479211395,
798
+ "learning_rate": 0.18227113921730334,
799
+ "loss": 1.3469,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.21923076923076923,
804
+ "grad_norm": 0.0049904735687066465,
805
+ "learning_rate": 0.1819152044288992,
806
+ "loss": 1.4624,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.22115384615384615,
811
+ "grad_norm": 0.004951369604388781,
812
+ "learning_rate": 0.18155608689592603,
813
+ "loss": 1.5918,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.2230769230769231,
818
+ "grad_norm": 0.0038554045596990603,
819
+ "learning_rate": 0.18119380057158568,
820
+ "loss": 1.5716,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.225,
825
+ "grad_norm": 0.004617105161027677,
826
+ "learning_rate": 0.18082835953220056,
827
+ "loss": 1.5557,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.22692307692307692,
832
+ "grad_norm": 0.0038371854835765997,
833
+ "learning_rate": 0.18045977797666685,
834
+ "loss": 1.4352,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.22884615384615384,
839
+ "grad_norm": 0.0043209401212792485,
840
+ "learning_rate": 0.1800880702259028,
841
+ "loss": 1.383,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.23076923076923078,
846
+ "grad_norm": 0.0048079513222180955,
847
+ "learning_rate": 0.17971325072229227,
848
+ "loss": 1.4152,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.2326923076923077,
853
+ "grad_norm": 0.005278957055411771,
854
+ "learning_rate": 0.17933533402912352,
855
+ "loss": 1.4748,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.23461538461538461,
860
+ "grad_norm": 0.003945878117350622,
861
+ "learning_rate": 0.17895433483002354,
862
+ "loss": 1.3498,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.23653846153846153,
867
+ "grad_norm": 0.006710115399694685,
868
+ "learning_rate": 0.17857026792838737,
869
+ "loss": 1.6258,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 0.23846153846153847,
874
+ "grad_norm": 0.004803140065677135,
875
+ "learning_rate": 0.178183148246803,
876
+ "loss": 1.4534,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 0.2403846153846154,
881
+ "grad_norm": 0.0048051798235238955,
882
+ "learning_rate": 0.1777929908264715,
883
+ "loss": 1.4264,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 0.2423076923076923,
888
+ "grad_norm": 0.003894997912593999,
889
+ "learning_rate": 0.17739981082662276,
890
+ "loss": 1.5164,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 0.24423076923076922,
895
+ "grad_norm": 0.005680190453774555,
896
+ "learning_rate": 0.1770036235239263,
897
+ "loss": 1.4138,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 0.24615384615384617,
902
+ "grad_norm": 0.004562874990462125,
903
+ "learning_rate": 0.1766044443118978,
904
+ "loss": 1.4616,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 0.24807692307692308,
909
+ "grad_norm": 0.0036123589153251607,
910
+ "learning_rate": 0.17620228870030108,
911
+ "loss": 1.3737,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 0.25,
916
+ "grad_norm": 0.004393834332318245,
917
+ "learning_rate": 0.1757971723145453,
918
+ "loss": 1.4444,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 0.2519230769230769,
923
+ "grad_norm": 0.005126787223383965,
924
+ "learning_rate": 0.175389110895078,
925
+ "loss": 1.4992,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 0.25384615384615383,
930
+ "grad_norm": 0.004323012600167896,
931
+ "learning_rate": 0.17497812029677343,
932
+ "loss": 1.4892,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 0.25576923076923075,
937
+ "grad_norm": 0.004101576559587375,
938
+ "learning_rate": 0.17456421648831655,
939
+ "loss": 1.389,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 0.25769230769230766,
944
+ "grad_norm": 0.004481754780758645,
945
+ "learning_rate": 0.17414741555158267,
946
+ "loss": 1.48,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 0.25961538461538464,
951
+ "grad_norm": 0.0039173182984178125,
952
+ "learning_rate": 0.1737277336810124,
953
+ "loss": 1.557,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 0.26153846153846155,
958
+ "grad_norm": 0.004081422317129897,
959
+ "learning_rate": 0.17330518718298263,
960
+ "loss": 1.4741,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 0.26346153846153847,
965
+ "grad_norm": 0.004979768820366094,
966
+ "learning_rate": 0.17287979247517285,
967
+ "loss": 1.3922,
968
+ "step": 137
969
+ },
970
+ {
971
+ "epoch": 0.2653846153846154,
972
+ "grad_norm": 0.003856969864870118,
973
+ "learning_rate": 0.17245156608592727,
974
+ "loss": 1.3935,
975
+ "step": 138
976
+ },
977
+ {
978
+ "epoch": 0.2673076923076923,
979
+ "grad_norm": 0.004262542287660523,
980
+ "learning_rate": 0.17202052465361267,
981
+ "loss": 1.3863,
982
+ "step": 139
983
+ },
984
+ {
985
+ "epoch": 0.2692307692307692,
986
+ "grad_norm": 0.004072321122248585,
987
+ "learning_rate": 0.17158668492597184,
988
+ "loss": 1.5358,
989
+ "step": 140
990
+ },
991
+ {
992
+ "epoch": 0.27115384615384613,
993
+ "grad_norm": 0.004129536323739746,
994
+ "learning_rate": 0.17115006375947303,
995
+ "loss": 1.5204,
996
+ "step": 141
997
+ },
998
+ {
999
+ "epoch": 0.27307692307692305,
1000
+ "grad_norm": 0.004315823383064392,
1001
+ "learning_rate": 0.17071067811865476,
1002
+ "loss": 1.5745,
1003
+ "step": 142
1004
+ },
1005
+ {
1006
+ "epoch": 0.275,
1007
+ "grad_norm": 0.004472884067435373,
1008
+ "learning_rate": 0.17026854507546693,
1009
+ "loss": 1.4347,
1010
+ "step": 143
1011
+ },
1012
+ {
1013
+ "epoch": 0.27692307692307694,
1014
+ "grad_norm": 0.004270920674869002,
1015
+ "learning_rate": 0.1698236818086073,
1016
+ "loss": 1.385,
1017
+ "step": 144
1018
+ },
1019
+ {
1020
+ "epoch": 0.27884615384615385,
1021
+ "grad_norm": 0.00392777927587438,
1022
+ "learning_rate": 0.16937610560285418,
1023
+ "loss": 1.3137,
1024
+ "step": 145
1025
+ },
1026
+ {
1027
+ "epoch": 0.28076923076923077,
1028
+ "grad_norm": 0.004459985692840538,
1029
+ "learning_rate": 0.1689258338483947,
1030
+ "loss": 1.6202,
1031
+ "step": 146
1032
+ },
1033
+ {
1034
+ "epoch": 0.2826923076923077,
1035
+ "grad_norm": 0.004514228785087149,
1036
+ "learning_rate": 0.16847288404014937,
1037
+ "loss": 1.3617,
1038
+ "step": 147
1039
+ },
1040
+ {
1041
+ "epoch": 0.2846153846153846,
1042
+ "grad_norm": 0.0038101679405580565,
1043
+ "learning_rate": 0.16801727377709194,
1044
+ "loss": 1.4024,
1045
+ "step": 148
1046
+ },
1047
+ {
1048
+ "epoch": 0.2865384615384615,
1049
+ "grad_norm": 0.0041582079409328105,
1050
+ "learning_rate": 0.16755902076156604,
1051
+ "loss": 1.3501,
1052
+ "step": 149
1053
+ },
1054
+ {
1055
+ "epoch": 0.28846153846153844,
1056
+ "grad_norm": 0.004383084305821378,
1057
+ "learning_rate": 0.16709814279859703,
1058
+ "loss": 1.5754,
1059
+ "step": 150
1060
+ },
1061
+ {
1062
+ "epoch": 0.2903846153846154,
1063
+ "grad_norm": 0.003769630388185354,
1064
+ "learning_rate": 0.1666346577952004,
1065
+ "loss": 1.3763,
1066
+ "step": 151
1067
+ },
1068
+ {
1069
+ "epoch": 0.2923076923076923,
1070
+ "grad_norm": 0.004322757794103019,
1071
+ "learning_rate": 0.16616858375968596,
1072
+ "loss": 1.3504,
1073
+ "step": 152
1074
+ },
1075
+ {
1076
+ "epoch": 0.29423076923076924,
1077
+ "grad_norm": 0.00451830897435612,
1078
+ "learning_rate": 0.16569993880095807,
1079
+ "loss": 1.3912,
1080
+ "step": 153
1081
+ },
1082
+ {
1083
+ "epoch": 0.29615384615384616,
1084
+ "grad_norm": 0.0037829375694366993,
1085
+ "learning_rate": 0.16522874112781213,
1086
+ "loss": 1.4656,
1087
+ "step": 154
1088
+ },
1089
+ {
1090
+ "epoch": 0.2980769230769231,
1091
+ "grad_norm": 0.004075462356384432,
1092
+ "learning_rate": 0.16475500904822704,
1093
+ "loss": 1.3765,
1094
+ "step": 155
1095
+ },
1096
+ {
1097
+ "epoch": 0.3,
1098
+ "grad_norm": 0.0045329044957026785,
1099
+ "learning_rate": 0.16427876096865393,
1100
+ "loss": 1.4106,
1101
+ "step": 156
1102
+ },
1103
+ {
1104
+ "epoch": 0.3019230769230769,
1105
+ "grad_norm": 0.004497276339013321,
1106
+ "learning_rate": 0.16380001539330089,
1107
+ "loss": 1.619,
1108
+ "step": 157
1109
+ },
1110
+ {
1111
+ "epoch": 0.3038461538461538,
1112
+ "grad_norm": 0.005291047147597019,
1113
+ "learning_rate": 0.163318790923414,
1114
+ "loss": 1.3879,
1115
+ "step": 158
1116
+ },
1117
+ {
1118
+ "epoch": 0.3057692307692308,
1119
+ "grad_norm": 0.0038260601439996204,
1120
+ "learning_rate": 0.16283510625655473,
1121
+ "loss": 1.4207,
1122
+ "step": 159
1123
+ },
1124
+ {
1125
+ "epoch": 0.3076923076923077,
1126
+ "grad_norm": 0.004192287890128624,
1127
+ "learning_rate": 0.16234898018587338,
1128
+ "loss": 1.4564,
1129
+ "step": 160
1130
+ },
1131
+ {
1132
+ "epoch": 0.3096153846153846,
1133
+ "grad_norm": 0.004321742014288,
1134
+ "learning_rate": 0.16186043159937882,
1135
+ "loss": 1.4293,
1136
+ "step": 161
1137
+ },
1138
+ {
1139
+ "epoch": 0.31153846153846154,
1140
+ "grad_norm": 0.004482133013951244,
1141
+ "learning_rate": 0.16136947947920477,
1142
+ "loss": 1.536,
1143
+ "step": 162
1144
+ },
1145
+ {
1146
+ "epoch": 0.31346153846153846,
1147
+ "grad_norm": 0.005158245749013815,
1148
+ "learning_rate": 0.16087614290087207,
1149
+ "loss": 1.3,
1150
+ "step": 163
1151
+ },
1152
+ {
1153
+ "epoch": 0.3153846153846154,
1154
+ "grad_norm": 0.0038679926592576986,
1155
+ "learning_rate": 0.16038044103254775,
1156
+ "loss": 1.2625,
1157
+ "step": 164
1158
+ },
1159
+ {
1160
+ "epoch": 0.3173076923076923,
1161
+ "grad_norm": 0.003840913634145776,
1162
+ "learning_rate": 0.15988239313430005,
1163
+ "loss": 1.4159,
1164
+ "step": 165
1165
+ },
1166
+ {
1167
+ "epoch": 0.3192307692307692,
1168
+ "grad_norm": 0.004438164835550007,
1169
+ "learning_rate": 0.15938201855735015,
1170
+ "loss": 1.4081,
1171
+ "step": 166
1172
+ },
1173
+ {
1174
+ "epoch": 0.3211538461538462,
1175
+ "grad_norm": 0.004432984083884321,
1176
+ "learning_rate": 0.15887933674332047,
1177
+ "loss": 1.4015,
1178
+ "step": 167
1179
+ },
1180
+ {
1181
+ "epoch": 0.3230769230769231,
1182
+ "grad_norm": 0.003944232086034696,
1183
+ "learning_rate": 0.158374367223479,
1184
+ "loss": 1.3339,
1185
+ "step": 168
1186
+ },
1187
+ {
1188
+ "epoch": 0.325,
1189
+ "grad_norm": 0.004278349086214503,
1190
+ "learning_rate": 0.1578671296179806,
1191
+ "loss": 1.4123,
1192
+ "step": 169
1193
+ },
1194
+ {
1195
+ "epoch": 0.3269230769230769,
1196
+ "grad_norm": 0.004561432228264618,
1197
+ "learning_rate": 0.15735764363510463,
1198
+ "loss": 1.4437,
1199
+ "step": 170
1200
+ },
1201
+ {
1202
+ "epoch": 0.32884615384615384,
1203
+ "grad_norm": 0.004434712504116213,
1204
+ "learning_rate": 0.15684592907048925,
1205
+ "loss": 1.3531,
1206
+ "step": 171
1207
+ },
1208
+ {
1209
+ "epoch": 0.33076923076923076,
1210
+ "grad_norm": 0.004513068924087115,
1211
+ "learning_rate": 0.1563320058063622,
1212
+ "loss": 1.4174,
1213
+ "step": 172
1214
+ },
1215
+ {
1216
+ "epoch": 0.3326923076923077,
1217
+ "grad_norm": 0.0038597641807620936,
1218
+ "learning_rate": 0.15581589381076844,
1219
+ "loss": 1.3457,
1220
+ "step": 173
1221
+ },
1222
+ {
1223
+ "epoch": 0.3346153846153846,
1224
+ "grad_norm": 0.00467448756637936,
1225
+ "learning_rate": 0.15529761313679394,
1226
+ "loss": 1.4238,
1227
+ "step": 174
1228
+ },
1229
+ {
1230
+ "epoch": 0.33653846153846156,
1231
+ "grad_norm": 0.00393730584602868,
1232
+ "learning_rate": 0.15477718392178716,
1233
+ "loss": 1.3146,
1234
+ "step": 175
1235
+ },
1236
+ {
1237
+ "epoch": 0.3384615384615385,
1238
+ "grad_norm": 0.004054159958305585,
1239
+ "learning_rate": 0.15425462638657594,
1240
+ "loss": 1.5487,
1241
+ "step": 176
1242
+ },
1243
+ {
1244
+ "epoch": 0.3403846153846154,
1245
+ "grad_norm": 0.004344506978623121,
1246
+ "learning_rate": 0.1537299608346824,
1247
+ "loss": 1.4104,
1248
+ "step": 177
1249
+ },
1250
+ {
1251
+ "epoch": 0.3423076923076923,
1252
+ "grad_norm": 0.004416572275984062,
1253
+ "learning_rate": 0.15320320765153367,
1254
+ "loss": 1.3928,
1255
+ "step": 178
1256
+ },
1257
+ {
1258
+ "epoch": 0.34423076923076923,
1259
+ "grad_norm": 0.003756679198600096,
1260
+ "learning_rate": 0.15267438730367008,
1261
+ "loss": 1.4747,
1262
+ "step": 179
1263
+ },
1264
+ {
1265
+ "epoch": 0.34615384615384615,
1266
+ "grad_norm": 0.004407377086662261,
1267
+ "learning_rate": 0.1521435203379498,
1268
+ "loss": 1.39,
1269
+ "step": 180
1270
+ },
1271
+ {
1272
+ "epoch": 0.34807692307692306,
1273
+ "grad_norm": 0.003933982639380978,
1274
+ "learning_rate": 0.15161062738075068,
1275
+ "loss": 1.3529,
1276
+ "step": 181
1277
+ },
1278
+ {
1279
+ "epoch": 0.35,
1280
+ "grad_norm": 0.0041841419203471975,
1281
+ "learning_rate": 0.1510757291371686,
1282
+ "loss": 1.3789,
1283
+ "step": 182
1284
+ },
1285
+ {
1286
+ "epoch": 0.35192307692307695,
1287
+ "grad_norm": 0.004120152636219925,
1288
+ "learning_rate": 0.1505388463902131,
1289
+ "loss": 1.3998,
1290
+ "step": 183
1291
+ },
1292
+ {
1293
+ "epoch": 0.35384615384615387,
1294
+ "grad_norm": 0.004140296087315911,
1295
+ "learning_rate": 0.15000000000000002,
1296
+ "loss": 1.3129,
1297
+ "step": 184
1298
+ },
1299
+ {
1300
+ "epoch": 0.3557692307692308,
1301
+ "grad_norm": 0.0038068591455658546,
1302
+ "learning_rate": 0.14945921090294076,
1303
+ "loss": 1.4985,
1304
+ "step": 185
1305
+ },
1306
+ {
1307
+ "epoch": 0.3576923076923077,
1308
+ "grad_norm": 0.004911975742467501,
1309
+ "learning_rate": 0.14891650011092894,
1310
+ "loss": 1.327,
1311
+ "step": 186
1312
+ },
1313
+ {
1314
+ "epoch": 0.3596153846153846,
1315
+ "grad_norm": 0.0050956256163564575,
1316
+ "learning_rate": 0.14837188871052398,
1317
+ "loss": 1.3413,
1318
+ "step": 187
1319
+ },
1320
+ {
1321
+ "epoch": 0.36153846153846153,
1322
+ "grad_norm": 0.003966974177502509,
1323
+ "learning_rate": 0.14782539786213184,
1324
+ "loss": 1.4205,
1325
+ "step": 188
1326
+ },
1327
+ {
1328
+ "epoch": 0.36346153846153845,
1329
+ "grad_norm": 0.003623574154435322,
1330
+ "learning_rate": 0.1472770487991827,
1331
+ "loss": 1.4408,
1332
+ "step": 189
1333
+ },
1334
+ {
1335
+ "epoch": 0.36538461538461536,
1336
+ "grad_norm": 0.004106470407105877,
1337
+ "learning_rate": 0.1467268628273062,
1338
+ "loss": 1.353,
1339
+ "step": 190
1340
+ },
1341
+ {
1342
+ "epoch": 0.36730769230769234,
1343
+ "grad_norm": 0.003690145560199488,
1344
+ "learning_rate": 0.1461748613235034,
1345
+ "loss": 1.2951,
1346
+ "step": 191
1347
+ },
1348
+ {
1349
+ "epoch": 0.36923076923076925,
1350
+ "grad_norm": 0.003895835372754922,
1351
+ "learning_rate": 0.1456210657353163,
1352
+ "loss": 1.4059,
1353
+ "step": 192
1354
+ },
1355
+ {
1356
+ "epoch": 0.37115384615384617,
1357
+ "grad_norm": 0.004482077528900326,
1358
+ "learning_rate": 0.14506549757999454,
1359
+ "loss": 1.4711,
1360
+ "step": 193
1361
+ },
1362
+ {
1363
+ "epoch": 0.3730769230769231,
1364
+ "grad_norm": 0.005377584827567655,
1365
+ "learning_rate": 0.14450817844365924,
1366
+ "loss": 1.3428,
1367
+ "step": 194
1368
+ },
1369
+ {
1370
+ "epoch": 0.375,
1371
+ "grad_norm": 0.003809996615412003,
1372
+ "learning_rate": 0.1439491299804645,
1373
+ "loss": 1.4185,
1374
+ "step": 195
1375
+ },
1376
+ {
1377
+ "epoch": 0.3769230769230769,
1378
+ "grad_norm": 0.004378911274012511,
1379
+ "learning_rate": 0.14338837391175582,
1380
+ "loss": 1.3809,
1381
+ "step": 196
1382
+ },
1383
+ {
1384
+ "epoch": 0.37884615384615383,
1385
+ "grad_norm": 0.0039685308046407235,
1386
+ "learning_rate": 0.14282593202522628,
1387
+ "loss": 1.3387,
1388
+ "step": 197
1389
+ },
1390
+ {
1391
+ "epoch": 0.38076923076923075,
1392
+ "grad_norm": 0.003985049521159734,
1393
+ "learning_rate": 0.14226182617406996,
1394
+ "loss": 1.414,
1395
+ "step": 198
1396
+ },
1397
+ {
1398
+ "epoch": 0.38269230769230766,
1399
+ "grad_norm": 0.0037323058518405985,
1400
+ "learning_rate": 0.14169607827613281,
1401
+ "loss": 1.3301,
1402
+ "step": 199
1403
+ },
1404
+ {
1405
+ "epoch": 0.38461538461538464,
1406
+ "grad_norm": 0.00453153369462826,
1407
+ "learning_rate": 0.14112871031306118,
1408
+ "loss": 1.4128,
1409
+ "step": 200
1410
+ },
1411
+ {
1412
+ "epoch": 0.38653846153846155,
1413
+ "grad_norm": 0.0035865405454796485,
1414
+ "learning_rate": 0.1405597443294475,
1415
+ "loss": 1.4204,
1416
+ "step": 201
1417
+ },
1418
+ {
1419
+ "epoch": 0.38846153846153847,
1420
+ "grad_norm": 0.003797714699282732,
1421
+ "learning_rate": 0.13998920243197407,
1422
+ "loss": 1.3252,
1423
+ "step": 202
1424
+ },
1425
+ {
1426
+ "epoch": 0.3903846153846154,
1427
+ "grad_norm": 0.003977612360306275,
1428
+ "learning_rate": 0.13941710678855396,
1429
+ "loss": 1.3827,
1430
+ "step": 203
1431
+ },
1432
+ {
1433
+ "epoch": 0.3923076923076923,
1434
+ "grad_norm": 0.004343581712140305,
1435
+ "learning_rate": 0.13884347962746948,
1436
+ "loss": 1.4162,
1437
+ "step": 204
1438
+ },
1439
+ {
1440
+ "epoch": 0.3942307692307692,
1441
+ "grad_norm": 0.004291027717547065,
1442
+ "learning_rate": 0.138268343236509,
1443
+ "loss": 1.4349,
1444
+ "step": 205
1445
+ },
1446
+ {
1447
+ "epoch": 0.39615384615384613,
1448
+ "grad_norm": 0.003966730709827027,
1449
+ "learning_rate": 0.13769171996210053,
1450
+ "loss": 1.4515,
1451
+ "step": 206
1452
+ },
1453
+ {
1454
+ "epoch": 0.39807692307692305,
1455
+ "grad_norm": 0.004433164664666871,
1456
+ "learning_rate": 0.1371136322084438,
1457
+ "loss": 1.4078,
1458
+ "step": 207
1459
+ },
1460
+ {
1461
+ "epoch": 0.4,
1462
+ "grad_norm": 0.0041301845311211205,
1463
+ "learning_rate": 0.13653410243663952,
1464
+ "loss": 1.4135,
1465
+ "step": 208
1466
+ },
1467
+ {
1468
+ "epoch": 0.40192307692307694,
1469
+ "grad_norm": 0.004007983656639604,
1470
+ "learning_rate": 0.13595315316381676,
1471
+ "loss": 1.3415,
1472
+ "step": 209
1473
+ },
1474
+ {
1475
+ "epoch": 0.40384615384615385,
1476
+ "grad_norm": 0.004364104237399665,
1477
+ "learning_rate": 0.13537080696225814,
1478
+ "loss": 1.422,
1479
+ "step": 210
1480
+ },
1481
+ {
1482
+ "epoch": 0.40576923076923077,
1483
+ "grad_norm": 0.0034359554102698768,
1484
+ "learning_rate": 0.13478708645852272,
1485
+ "loss": 1.4224,
1486
+ "step": 211
1487
+ },
1488
+ {
1489
+ "epoch": 0.4076923076923077,
1490
+ "grad_norm": 0.003665490745319788,
1491
+ "learning_rate": 0.1342020143325669,
1492
+ "loss": 1.3919,
1493
+ "step": 212
1494
+ },
1495
+ {
1496
+ "epoch": 0.4096153846153846,
1497
+ "grad_norm": 0.004823134122585627,
1498
+ "learning_rate": 0.13361561331686309,
1499
+ "loss": 1.354,
1500
+ "step": 213
1501
+ },
1502
+ {
1503
+ "epoch": 0.4115384615384615,
1504
+ "grad_norm": 0.004164134706719053,
1505
+ "learning_rate": 0.13302790619551672,
1506
+ "loss": 1.3459,
1507
+ "step": 214
1508
+ },
1509
+ {
1510
+ "epoch": 0.41346153846153844,
1511
+ "grad_norm": 0.0038419796032502142,
1512
+ "learning_rate": 0.1324389158033807,
1513
+ "loss": 1.3407,
1514
+ "step": 215
1515
+ },
1516
+ {
1517
+ "epoch": 0.4153846153846154,
1518
+ "grad_norm": 0.003684341396676353,
1519
+ "learning_rate": 0.13184866502516845,
1520
+ "loss": 1.2564,
1521
+ "step": 216
1522
+ },
1523
+ {
1524
+ "epoch": 0.4173076923076923,
1525
+ "grad_norm": 0.0037530100188813984,
1526
+ "learning_rate": 0.13125717679456447,
1527
+ "loss": 1.3893,
1528
+ "step": 217
1529
+ },
1530
+ {
1531
+ "epoch": 0.41923076923076924,
1532
+ "grad_norm": 0.003871346852120206,
1533
+ "learning_rate": 0.13066447409333345,
1534
+ "loss": 1.3799,
1535
+ "step": 218
1536
+ },
1537
+ {
1538
+ "epoch": 0.42115384615384616,
1539
+ "grad_norm": 0.003333692515234481,
1540
+ "learning_rate": 0.1300705799504273,
1541
+ "loss": 1.3438,
1542
+ "step": 219
1543
+ },
1544
+ {
1545
+ "epoch": 0.4230769230769231,
1546
+ "grad_norm": 0.0037740383957130296,
1547
+ "learning_rate": 0.12947551744109043,
1548
+ "loss": 1.3856,
1549
+ "step": 220
1550
+ },
1551
+ {
1552
+ "epoch": 0.425,
1553
+ "grad_norm": 0.0038981706525231356,
1554
+ "learning_rate": 0.128879309685963,
1555
+ "loss": 1.3842,
1556
+ "step": 221
1557
+ },
1558
+ {
1559
+ "epoch": 0.4269230769230769,
1560
+ "grad_norm": 0.00351449664077274,
1561
+ "learning_rate": 0.12828197985018275,
1562
+ "loss": 1.2889,
1563
+ "step": 222
1564
+ },
1565
+ {
1566
+ "epoch": 0.4288461538461538,
1567
+ "grad_norm": 0.0038545081864729855,
1568
+ "learning_rate": 0.12768355114248495,
1569
+ "loss": 1.2847,
1570
+ "step": 223
1571
+ },
1572
+ {
1573
+ "epoch": 0.4307692307692308,
1574
+ "grad_norm": 0.005942469596135689,
1575
+ "learning_rate": 0.12708404681430052,
1576
+ "loss": 1.583,
1577
+ "step": 224
1578
+ },
1579
+ {
1580
+ "epoch": 0.4326923076923077,
1581
+ "grad_norm": 0.004015959113750885,
1582
+ "learning_rate": 0.1264834901588527,
1583
+ "loss": 1.3066,
1584
+ "step": 225
1585
+ },
1586
+ {
1587
+ "epoch": 0.4346153846153846,
1588
+ "grad_norm": 0.0036938637774224423,
1589
+ "learning_rate": 0.12588190451025208,
1590
+ "loss": 1.4096,
1591
+ "step": 226
1592
+ },
1593
+ {
1594
+ "epoch": 0.43653846153846154,
1595
+ "grad_norm": 0.0035132148158606006,
1596
+ "learning_rate": 0.12527931324258976,
1597
+ "loss": 1.3859,
1598
+ "step": 227
1599
+ },
1600
+ {
1601
+ "epoch": 0.43846153846153846,
1602
+ "grad_norm": 0.004399808865437533,
1603
+ "learning_rate": 0.12467573976902935,
1604
+ "loss": 1.5347,
1605
+ "step": 228
1606
+ },
1607
+ {
1608
+ "epoch": 0.4403846153846154,
1609
+ "grad_norm": 0.0036096537861978212,
1610
+ "learning_rate": 0.12407120754089732,
1611
+ "loss": 1.3746,
1612
+ "step": 229
1613
+ },
1614
+ {
1615
+ "epoch": 0.4423076923076923,
1616
+ "grad_norm": 0.0038518788469343213,
1617
+ "learning_rate": 0.12346574004677154,
1618
+ "loss": 1.2391,
1619
+ "step": 230
1620
+ },
1621
+ {
1622
+ "epoch": 0.4442307692307692,
1623
+ "grad_norm": 0.0033816738538052084,
1624
+ "learning_rate": 0.12285936081156897,
1625
+ "loss": 1.304,
1626
+ "step": 231
1627
+ },
1628
+ {
1629
+ "epoch": 0.4461538461538462,
1630
+ "grad_norm": 0.004038981298220094,
1631
+ "learning_rate": 0.12225209339563144,
1632
+ "loss": 1.5596,
1633
+ "step": 232
1634
+ },
1635
+ {
1636
+ "epoch": 0.4480769230769231,
1637
+ "grad_norm": 0.004622913540091291,
1638
+ "learning_rate": 0.12164396139381028,
1639
+ "loss": 1.4443,
1640
+ "step": 233
1641
+ },
1642
+ {
1643
+ "epoch": 0.45,
1644
+ "grad_norm": 0.0038598204714652335,
1645
+ "learning_rate": 0.12103498843454959,
1646
+ "loss": 1.2573,
1647
+ "step": 234
1648
+ },
1649
+ {
1650
+ "epoch": 0.4519230769230769,
1651
+ "grad_norm": 0.004201667715738951,
1652
+ "learning_rate": 0.12042519817896805,
1653
+ "loss": 1.3096,
1654
+ "step": 235
1655
+ },
1656
+ {
1657
+ "epoch": 0.45384615384615384,
1658
+ "grad_norm": 0.003477755766612118,
1659
+ "learning_rate": 0.11981461431993977,
1660
+ "loss": 1.4307,
1661
+ "step": 236
1662
+ },
1663
+ {
1664
+ "epoch": 0.45576923076923076,
1665
+ "grad_norm": 0.0036754779197291565,
1666
+ "learning_rate": 0.11920326058117364,
1667
+ "loss": 1.3851,
1668
+ "step": 237
1669
+ },
1670
+ {
1671
+ "epoch": 0.4576923076923077,
1672
+ "grad_norm": 0.0037609693641422583,
1673
+ "learning_rate": 0.11859116071629149,
1674
+ "loss": 1.33,
1675
+ "step": 238
1676
+ },
1677
+ {
1678
+ "epoch": 0.4596153846153846,
1679
+ "grad_norm": 0.0039586101696845516,
1680
+ "learning_rate": 0.11797833850790528,
1681
+ "loss": 1.4323,
1682
+ "step": 239
1683
+ },
1684
+ {
1685
+ "epoch": 0.46153846153846156,
1686
+ "grad_norm": 0.004000335218972953,
1687
+ "learning_rate": 0.11736481776669305,
1688
+ "loss": 1.1934,
1689
+ "step": 240
1690
+ },
1691
+ {
1692
+ "epoch": 0.4634615384615385,
1693
+ "grad_norm": 0.0038820374980957696,
1694
+ "learning_rate": 0.11675062233047365,
1695
+ "loss": 1.2851,
1696
+ "step": 241
1697
+ },
1698
+ {
1699
+ "epoch": 0.4653846153846154,
1700
+ "grad_norm": 0.0035547528827412204,
1701
+ "learning_rate": 0.11613577606328068,
1702
+ "loss": 1.3022,
1703
+ "step": 242
1704
+ },
1705
+ {
1706
+ "epoch": 0.4673076923076923,
1707
+ "grad_norm": 0.00361234070971243,
1708
+ "learning_rate": 0.11552030285443515,
1709
+ "loss": 1.2865,
1710
+ "step": 243
1711
+ },
1712
+ {
1713
+ "epoch": 0.46923076923076923,
1714
+ "grad_norm": 0.003835737992005575,
1715
+ "learning_rate": 0.11490422661761744,
1716
+ "loss": 1.432,
1717
+ "step": 244
1718
+ },
1719
+ {
1720
+ "epoch": 0.47115384615384615,
1721
+ "grad_norm": 0.0036502184290390116,
1722
+ "learning_rate": 0.11428757128993801,
1723
+ "loss": 1.2768,
1724
+ "step": 245
1725
+ },
1726
+ {
1727
+ "epoch": 0.47307692307692306,
1728
+ "grad_norm": 0.004233011945267489,
1729
+ "learning_rate": 0.11367036083100734,
1730
+ "loss": 1.5403,
1731
+ "step": 246
1732
+ },
1733
+ {
1734
+ "epoch": 0.475,
1735
+ "grad_norm": 0.0038626796874710275,
1736
+ "learning_rate": 0.11305261922200519,
1737
+ "loss": 1.4572,
1738
+ "step": 247
1739
+ },
1740
+ {
1741
+ "epoch": 0.47692307692307695,
1742
+ "grad_norm": 0.0036737039608909275,
1743
+ "learning_rate": 0.11243437046474854,
1744
+ "loss": 1.2777,
1745
+ "step": 248
1746
+ },
1747
+ {
1748
+ "epoch": 0.47884615384615387,
1749
+ "grad_norm": 0.0038629787695078177,
1750
+ "learning_rate": 0.1118156385807593,
1751
+ "loss": 1.3932,
1752
+ "step": 249
1753
+ },
1754
+ {
1755
+ "epoch": 0.4807692307692308,
1756
+ "grad_norm": 0.004213134602869909,
1757
+ "learning_rate": 0.11119644761033079,
1758
+ "loss": 1.3311,
1759
+ "step": 250
1760
+ },
1761
+ {
1762
+ "epoch": 0.4826923076923077,
1763
+ "grad_norm": 0.0034652845886843594,
1764
+ "learning_rate": 0.1105768216115938,
1765
+ "loss": 1.3913,
1766
+ "step": 251
1767
+ },
1768
+ {
1769
+ "epoch": 0.4846153846153846,
1770
+ "grad_norm": 0.003815549252597903,
1771
+ "learning_rate": 0.10995678465958168,
1772
+ "loss": 1.4117,
1773
+ "step": 252
1774
+ },
1775
+ {
1776
+ "epoch": 0.48653846153846153,
1777
+ "grad_norm": 0.003951861516294408,
1778
+ "learning_rate": 0.10933636084529506,
1779
+ "loss": 1.3874,
1780
+ "step": 253
1781
+ },
1782
+ {
1783
+ "epoch": 0.48846153846153845,
1784
+ "grad_norm": 0.003357731090297942,
1785
+ "learning_rate": 0.10871557427476584,
1786
+ "loss": 1.2948,
1787
+ "step": 254
1788
+ },
1789
+ {
1790
+ "epoch": 0.49038461538461536,
1791
+ "grad_norm": 0.003831966087879381,
1792
+ "learning_rate": 0.10809444906812034,
1793
+ "loss": 1.3169,
1794
+ "step": 255
1795
+ },
1796
+ {
1797
+ "epoch": 0.49230769230769234,
1798
+ "grad_norm": 0.004106166707429391,
1799
+ "learning_rate": 0.10747300935864244,
1800
+ "loss": 1.361,
1801
+ "step": 256
1802
+ },
1803
+ {
1804
+ "epoch": 0.49423076923076925,
1805
+ "grad_norm": 0.004111767769905025,
1806
+ "learning_rate": 0.10685127929183567,
1807
+ "loss": 1.3509,
1808
+ "step": 257
1809
+ },
1810
+ {
1811
+ "epoch": 0.49615384615384617,
1812
+ "grad_norm": 0.003261641190292533,
1813
+ "learning_rate": 0.10622928302448523,
1814
+ "loss": 1.3704,
1815
+ "step": 258
1816
+ },
1817
+ {
1818
+ "epoch": 0.4980769230769231,
1819
+ "grad_norm": 0.004090034168359905,
1820
+ "learning_rate": 0.10560704472371918,
1821
+ "loss": 1.4313,
1822
+ "step": 259
1823
+ },
1824
+ {
1825
+ "epoch": 0.5,
1826
+ "grad_norm": 0.0036280366628641592,
1827
+ "learning_rate": 0.10498458856606972,
1828
+ "loss": 1.4976,
1829
+ "step": 260
1830
+ },
1831
+ {
1832
+ "epoch": 0.5019230769230769,
1833
+ "grad_norm": 0.0037335189907437252,
1834
+ "learning_rate": 0.10436193873653361,
1835
+ "loss": 1.4199,
1836
+ "step": 261
1837
+ },
1838
+ {
1839
+ "epoch": 0.5038461538461538,
1840
+ "grad_norm": 0.003742267177907334,
1841
+ "learning_rate": 0.10373911942763259,
1842
+ "loss": 1.275,
1843
+ "step": 262
1844
+ },
1845
+ {
1846
+ "epoch": 0.5057692307692307,
1847
+ "grad_norm": 0.0037564481395607778,
1848
+ "learning_rate": 0.10311615483847332,
1849
+ "loss": 1.4365,
1850
+ "step": 263
1851
+ },
1852
+ {
1853
+ "epoch": 0.5076923076923077,
1854
+ "grad_norm": 0.0035504233601682273,
1855
+ "learning_rate": 0.1024930691738073,
1856
+ "loss": 1.3999,
1857
+ "step": 264
1858
+ },
1859
+ {
1860
+ "epoch": 0.5096153846153846,
1861
+ "grad_norm": 0.004089383910203458,
1862
+ "learning_rate": 0.10186988664309023,
1863
+ "loss": 1.2874,
1864
+ "step": 265
1865
+ },
1866
+ {
1867
+ "epoch": 0.5115384615384615,
1868
+ "grad_norm": 0.003152687093182755,
1869
+ "learning_rate": 0.10124663145954152,
1870
+ "loss": 1.133,
1871
+ "step": 266
1872
+ },
1873
+ {
1874
+ "epoch": 0.5134615384615384,
1875
+ "grad_norm": 0.003549991155156565,
1876
+ "learning_rate": 0.10062332783920336,
1877
+ "loss": 1.2866,
1878
+ "step": 267
1879
+ },
1880
+ {
1881
+ "epoch": 0.5153846153846153,
1882
+ "grad_norm": 0.00426112090456259,
1883
+ "learning_rate": 0.1,
1884
+ "loss": 1.5642,
1885
+ "step": 268
1886
+ },
1887
+ {
1888
+ "epoch": 0.5173076923076924,
1889
+ "grad_norm": 0.0037774022022750552,
1890
+ "learning_rate": 0.09937667216079665,
1891
+ "loss": 1.4039,
1892
+ "step": 269
1893
+ },
1894
+ {
1895
+ "epoch": 0.5192307692307693,
1896
+ "grad_norm": 0.0037180769854636055,
1897
+ "learning_rate": 0.0987533685404585,
1898
+ "loss": 1.3444,
1899
+ "step": 270
1900
+ },
1901
+ {
1902
+ "epoch": 0.5211538461538462,
1903
+ "grad_norm": 0.0037451233297631646,
1904
+ "learning_rate": 0.0981301133569098,
1905
+ "loss": 1.3828,
1906
+ "step": 271
1907
+ },
1908
+ {
1909
+ "epoch": 0.5230769230769231,
1910
+ "grad_norm": 0.004023528701778675,
1911
+ "learning_rate": 0.09750693082619273,
1912
+ "loss": 1.3517,
1913
+ "step": 272
1914
+ },
1915
+ {
1916
+ "epoch": 0.525,
1917
+ "grad_norm": 0.003902891767888948,
1918
+ "learning_rate": 0.0968838451615267,
1919
+ "loss": 1.4862,
1920
+ "step": 273
1921
+ },
1922
+ {
1923
+ "epoch": 0.5269230769230769,
1924
+ "grad_norm": 0.003976022495281997,
1925
+ "learning_rate": 0.09626088057236745,
1926
+ "loss": 1.341,
1927
+ "step": 274
1928
+ },
1929
+ {
1930
+ "epoch": 0.5288461538461539,
1931
+ "grad_norm": 0.004041517649118537,
1932
+ "learning_rate": 0.09563806126346641,
1933
+ "loss": 1.2873,
1934
+ "step": 275
1935
+ },
1936
+ {
1937
+ "epoch": 0.5307692307692308,
1938
+ "grad_norm": 0.004195418934576378,
1939
+ "learning_rate": 0.09501541143393027,
1940
+ "loss": 1.3697,
1941
+ "step": 276
1942
+ },
1943
+ {
1944
+ "epoch": 0.5326923076923077,
1945
+ "grad_norm": 0.004400264773170083,
1946
+ "learning_rate": 0.09439295527628082,
1947
+ "loss": 1.4761,
1948
+ "step": 277
1949
+ },
1950
+ {
1951
+ "epoch": 0.5346153846153846,
1952
+ "grad_norm": 0.0034470003000413585,
1953
+ "learning_rate": 0.0937707169755148,
1954
+ "loss": 1.238,
1955
+ "step": 278
1956
+ },
1957
+ {
1958
+ "epoch": 0.5365384615384615,
1959
+ "grad_norm": 0.004485919966435778,
1960
+ "learning_rate": 0.09314872070816434,
1961
+ "loss": 1.3739,
1962
+ "step": 279
1963
+ },
1964
+ {
1965
+ "epoch": 0.5384615384615384,
1966
+ "grad_norm": 0.004215256949958898,
1967
+ "learning_rate": 0.09252699064135758,
1968
+ "loss": 1.2962,
1969
+ "step": 280
1970
+ },
1971
+ {
1972
+ "epoch": 0.5403846153846154,
1973
+ "grad_norm": 0.0038625535701206208,
1974
+ "learning_rate": 0.09190555093187967,
1975
+ "loss": 1.4012,
1976
+ "step": 281
1977
+ },
1978
+ {
1979
+ "epoch": 0.5423076923076923,
1980
+ "grad_norm": 0.00343599856219453,
1981
+ "learning_rate": 0.09128442572523418,
1982
+ "loss": 1.2402,
1983
+ "step": 282
1984
+ },
1985
+ {
1986
+ "epoch": 0.5442307692307692,
1987
+ "grad_norm": 0.004313511295564648,
1988
+ "learning_rate": 0.09066363915470495,
1989
+ "loss": 1.4241,
1990
+ "step": 283
1991
+ },
1992
+ {
1993
+ "epoch": 0.5461538461538461,
1994
+ "grad_norm": 0.00442162131379324,
1995
+ "learning_rate": 0.09004321534041836,
1996
+ "loss": 1.3457,
1997
+ "step": 284
1998
+ },
1999
+ {
2000
+ "epoch": 0.5480769230769231,
2001
+ "grad_norm": 0.003915459812845989,
2002
+ "learning_rate": 0.08942317838840624,
2003
+ "loss": 1.2744,
2004
+ "step": 285
2005
+ },
2006
+ {
2007
+ "epoch": 0.55,
2008
+ "grad_norm": 0.004143308101542936,
2009
+ "learning_rate": 0.08880355238966922,
2010
+ "loss": 1.1394,
2011
+ "step": 286
2012
+ },
2013
+ {
2014
+ "epoch": 0.551923076923077,
2015
+ "grad_norm": 0.0040216024965500025,
2016
+ "learning_rate": 0.08818436141924073,
2017
+ "loss": 1.3976,
2018
+ "step": 287
2019
+ },
2020
+ {
2021
+ "epoch": 0.5538461538461539,
2022
+ "grad_norm": 0.003944983102980846,
2023
+ "learning_rate": 0.08756562953525152,
2024
+ "loss": 1.4441,
2025
+ "step": 288
2026
+ },
2027
+ {
2028
+ "epoch": 0.5557692307692308,
2029
+ "grad_norm": 0.0034626844283658286,
2030
+ "learning_rate": 0.08694738077799487,
2031
+ "loss": 1.2929,
2032
+ "step": 289
2033
+ },
2034
+ {
2035
+ "epoch": 0.5576923076923077,
2036
+ "grad_norm": 0.003406015789983572,
2037
+ "learning_rate": 0.08632963916899268,
2038
+ "loss": 1.2039,
2039
+ "step": 290
2040
+ },
2041
+ {
2042
+ "epoch": 0.5596153846153846,
2043
+ "grad_norm": 0.004071918905574612,
2044
+ "learning_rate": 0.08571242871006202,
2045
+ "loss": 1.2671,
2046
+ "step": 291
2047
+ },
2048
+ {
2049
+ "epoch": 0.5615384615384615,
2050
+ "grad_norm": 0.003437016252262741,
2051
+ "learning_rate": 0.08509577338238256,
2052
+ "loss": 1.3105,
2053
+ "step": 292
2054
+ },
2055
+ {
2056
+ "epoch": 0.5634615384615385,
2057
+ "grad_norm": 0.003804630756934808,
2058
+ "learning_rate": 0.08447969714556484,
2059
+ "loss": 1.2525,
2060
+ "step": 293
2061
+ },
2062
+ {
2063
+ "epoch": 0.5653846153846154,
2064
+ "grad_norm": 0.003892025412083441,
2065
+ "learning_rate": 0.08386422393671933,
2066
+ "loss": 1.2856,
2067
+ "step": 294
2068
+ },
2069
+ {
2070
+ "epoch": 0.5673076923076923,
2071
+ "grad_norm": 0.005074363684964378,
2072
+ "learning_rate": 0.08324937766952638,
2073
+ "loss": 1.4218,
2074
+ "step": 295
2075
+ },
2076
+ {
2077
+ "epoch": 0.5692307692307692,
2078
+ "grad_norm": 0.003867763144246398,
2079
+ "learning_rate": 0.08263518223330697,
2080
+ "loss": 1.2263,
2081
+ "step": 296
2082
+ },
2083
+ {
2084
+ "epoch": 0.5711538461538461,
2085
+ "grad_norm": 0.00394638125369802,
2086
+ "learning_rate": 0.08202166149209474,
2087
+ "loss": 1.3721,
2088
+ "step": 297
2089
+ },
2090
+ {
2091
+ "epoch": 0.573076923076923,
2092
+ "grad_norm": 0.0033766338956703816,
2093
+ "learning_rate": 0.08140883928370855,
2094
+ "loss": 1.3276,
2095
+ "step": 298
2096
+ },
2097
+ {
2098
+ "epoch": 0.575,
2099
+ "grad_norm": 0.0036326118333781065,
2100
+ "learning_rate": 0.0807967394188264,
2101
+ "loss": 1.428,
2102
+ "step": 299
2103
+ },
2104
+ {
2105
+ "epoch": 0.5769230769230769,
2106
+ "grad_norm": 0.0037026776415979303,
2107
+ "learning_rate": 0.08018538568006027,
2108
+ "loss": 1.3849,
2109
+ "step": 300
2110
+ },
2111
+ {
2112
+ "epoch": 0.5788461538461539,
2113
+ "grad_norm": 0.0036568805923053805,
2114
+ "learning_rate": 0.07957480182103199,
2115
+ "loss": 1.3554,
2116
+ "step": 301
2117
+ },
2118
+ {
2119
+ "epoch": 0.5807692307692308,
2120
+ "grad_norm": 0.0044598685366664045,
2121
+ "learning_rate": 0.07896501156545044,
2122
+ "loss": 1.4431,
2123
+ "step": 302
2124
+ },
2125
+ {
2126
+ "epoch": 0.5826923076923077,
2127
+ "grad_norm": 0.003899025162159858,
2128
+ "learning_rate": 0.07835603860618973,
2129
+ "loss": 1.2933,
2130
+ "step": 303
2131
+ },
2132
+ {
2133
+ "epoch": 0.5846153846153846,
2134
+ "grad_norm": 0.003952269200120353,
2135
+ "learning_rate": 0.07774790660436857,
2136
+ "loss": 1.3336,
2137
+ "step": 304
2138
+ },
2139
+ {
2140
+ "epoch": 0.5865384615384616,
2141
+ "grad_norm": 0.004113044536001592,
2142
+ "learning_rate": 0.07714063918843106,
2143
+ "loss": 1.4022,
2144
+ "step": 305
2145
+ },
2146
+ {
2147
+ "epoch": 0.5884615384615385,
2148
+ "grad_norm": 0.003577017242965219,
2149
+ "learning_rate": 0.0765342599532285,
2150
+ "loss": 1.3277,
2151
+ "step": 306
2152
+ },
2153
+ {
2154
+ "epoch": 0.5903846153846154,
2155
+ "grad_norm": 0.0035834793797846474,
2156
+ "learning_rate": 0.07592879245910272,
2157
+ "loss": 1.2712,
2158
+ "step": 307
2159
+ },
2160
+ {
2161
+ "epoch": 0.5923076923076923,
2162
+ "grad_norm": 0.0034827070213127647,
2163
+ "learning_rate": 0.07532426023097064,
2164
+ "loss": 1.3953,
2165
+ "step": 308
2166
+ },
2167
+ {
2168
+ "epoch": 0.5942307692307692,
2169
+ "grad_norm": 0.003374322671931844,
2170
+ "learning_rate": 0.07472068675741024,
2171
+ "loss": 1.2676,
2172
+ "step": 309
2173
+ },
2174
+ {
2175
+ "epoch": 0.5961538461538461,
2176
+ "grad_norm": 0.003646674712249218,
2177
+ "learning_rate": 0.07411809548974792,
2178
+ "loss": 1.2473,
2179
+ "step": 310
2180
+ },
2181
+ {
2182
+ "epoch": 0.5980769230769231,
2183
+ "grad_norm": 0.0035497204864493237,
2184
+ "learning_rate": 0.07351650984114727,
2185
+ "loss": 1.2192,
2186
+ "step": 311
2187
+ },
2188
+ {
2189
+ "epoch": 0.6,
2190
+ "grad_norm": 0.004069462202304509,
2191
+ "learning_rate": 0.0729159531856995,
2192
+ "loss": 1.2049,
2193
+ "step": 312
2194
+ },
2195
+ {
2196
+ "epoch": 0.6019230769230769,
2197
+ "grad_norm": 0.003318409494062338,
2198
+ "learning_rate": 0.07231644885751508,
2199
+ "loss": 1.2034,
2200
+ "step": 313
2201
+ },
2202
+ {
2203
+ "epoch": 0.6038461538461538,
2204
+ "grad_norm": 0.0034322806303290865,
2205
+ "learning_rate": 0.07171802014981725,
2206
+ "loss": 1.2352,
2207
+ "step": 314
2208
+ },
2209
+ {
2210
+ "epoch": 0.6057692307692307,
2211
+ "grad_norm": 0.004519815605605601,
2212
+ "learning_rate": 0.07112069031403703,
2213
+ "loss": 1.4042,
2214
+ "step": 315
2215
+ },
2216
+ {
2217
+ "epoch": 0.6076923076923076,
2218
+ "grad_norm": 0.004197854152714484,
2219
+ "learning_rate": 0.07052448255890957,
2220
+ "loss": 1.2065,
2221
+ "step": 316
2222
+ },
2223
+ {
2224
+ "epoch": 0.6096153846153847,
2225
+ "grad_norm": 0.003266474378814634,
2226
+ "learning_rate": 0.0699294200495727,
2227
+ "loss": 1.2321,
2228
+ "step": 317
2229
+ },
2230
+ {
2231
+ "epoch": 0.6115384615384616,
2232
+ "grad_norm": 0.003865675432633035,
2233
+ "learning_rate": 0.06933552590666658,
2234
+ "loss": 1.3589,
2235
+ "step": 318
2236
+ },
2237
+ {
2238
+ "epoch": 0.6134615384615385,
2239
+ "grad_norm": 0.003648481209361014,
2240
+ "learning_rate": 0.06874282320543557,
2241
+ "loss": 1.2195,
2242
+ "step": 319
2243
+ },
2244
+ {
2245
+ "epoch": 0.6153846153846154,
2246
+ "grad_norm": 0.0038770425729092407,
2247
+ "learning_rate": 0.06815133497483157,
2248
+ "loss": 1.1601,
2249
+ "step": 320
2250
+ },
2251
+ {
2252
+ "epoch": 0.6173076923076923,
2253
+ "grad_norm": 0.003975331764495512,
2254
+ "learning_rate": 0.06756108419661931,
2255
+ "loss": 1.3726,
2256
+ "step": 321
2257
+ },
2258
+ {
2259
+ "epoch": 0.6192307692307693,
2260
+ "grad_norm": 0.0037814571144303586,
2261
+ "learning_rate": 0.06697209380448332,
2262
+ "loss": 1.2579,
2263
+ "step": 322
2264
+ },
2265
+ {
2266
+ "epoch": 0.6211538461538462,
2267
+ "grad_norm": 0.004359813756884978,
2268
+ "learning_rate": 0.06638438668313694,
2269
+ "loss": 1.341,
2270
+ "step": 323
2271
+ },
2272
+ {
2273
+ "epoch": 0.6230769230769231,
2274
+ "grad_norm": 0.0045645157130493755,
2275
+ "learning_rate": 0.06579798566743314,
2276
+ "loss": 1.295,
2277
+ "step": 324
2278
+ },
2279
+ {
2280
+ "epoch": 0.625,
2281
+ "grad_norm": 0.003918126878459061,
2282
+ "learning_rate": 0.06521291354147728,
2283
+ "loss": 1.3135,
2284
+ "step": 325
2285
+ },
2286
+ {
2287
+ "epoch": 0.6269230769230769,
2288
+ "grad_norm": 0.0037040800008645586,
2289
+ "learning_rate": 0.06462919303774187,
2290
+ "loss": 1.2911,
2291
+ "step": 326
2292
+ },
2293
+ {
2294
+ "epoch": 0.6288461538461538,
2295
+ "grad_norm": 0.0045778203265612,
2296
+ "learning_rate": 0.06404684683618325,
2297
+ "loss": 1.4126,
2298
+ "step": 327
2299
+ },
2300
+ {
2301
+ "epoch": 0.6307692307692307,
2302
+ "grad_norm": 0.004171294032466974,
2303
+ "learning_rate": 0.0634658975633605,
2304
+ "loss": 1.3611,
2305
+ "step": 328
2306
+ },
2307
+ {
2308
+ "epoch": 0.6326923076923077,
2309
+ "grad_norm": 0.003182202721742736,
2310
+ "learning_rate": 0.06288636779155621,
2311
+ "loss": 1.2061,
2312
+ "step": 329
2313
+ },
2314
+ {
2315
+ "epoch": 0.6346153846153846,
2316
+ "grad_norm": 0.003286197838419992,
2317
+ "learning_rate": 0.06230828003789948,
2318
+ "loss": 1.2815,
2319
+ "step": 330
2320
+ },
2321
+ {
2322
+ "epoch": 0.6365384615384615,
2323
+ "grad_norm": 0.003491949017820312,
2324
+ "learning_rate": 0.06173165676349103,
2325
+ "loss": 1.25,
2326
+ "step": 331
2327
+ },
2328
+ {
2329
+ "epoch": 0.6384615384615384,
2330
+ "grad_norm": 0.003961179821701042,
2331
+ "learning_rate": 0.06115652037253053,
2332
+ "loss": 1.4123,
2333
+ "step": 332
2334
+ },
2335
+ {
2336
+ "epoch": 0.6403846153846153,
2337
+ "grad_norm": 0.004018491490934039,
2338
+ "learning_rate": 0.06058289321144608,
2339
+ "loss": 1.4145,
2340
+ "step": 333
2341
+ },
2342
+ {
2343
+ "epoch": 0.6423076923076924,
2344
+ "grad_norm": 0.004252044885407218,
2345
+ "learning_rate": 0.06001079756802592,
2346
+ "loss": 1.2977,
2347
+ "step": 334
2348
+ },
2349
+ {
2350
+ "epoch": 0.6442307692307693,
2351
+ "grad_norm": 0.00325579789007877,
2352
+ "learning_rate": 0.059440255670552514,
2353
+ "loss": 1.289,
2354
+ "step": 335
2355
+ },
2356
+ {
2357
+ "epoch": 0.6461538461538462,
2358
+ "grad_norm": 0.004047461168898875,
2359
+ "learning_rate": 0.05887128968693887,
2360
+ "loss": 1.1755,
2361
+ "step": 336
2362
+ },
2363
+ {
2364
+ "epoch": 0.6480769230769231,
2365
+ "grad_norm": 0.0037558393555422653,
2366
+ "learning_rate": 0.058303921723867225,
2367
+ "loss": 1.1795,
2368
+ "step": 337
2369
+ },
2370
+ {
2371
+ "epoch": 0.65,
2372
+ "grad_norm": 0.003620014344023806,
2373
+ "learning_rate": 0.05773817382593008,
2374
+ "loss": 1.3042,
2375
+ "step": 338
2376
+ },
2377
+ {
2378
+ "epoch": 0.6519230769230769,
2379
+ "grad_norm": 0.003564589802053869,
2380
+ "learning_rate": 0.057174067974773715,
2381
+ "loss": 1.2434,
2382
+ "step": 339
2383
+ },
2384
+ {
2385
+ "epoch": 0.6538461538461539,
2386
+ "grad_norm": 0.003630694027969668,
2387
+ "learning_rate": 0.056611626088244195,
2388
+ "loss": 1.2355,
2389
+ "step": 340
2390
+ },
2391
+ {
2392
+ "epoch": 0.6557692307692308,
2393
+ "grad_norm": 0.003879005458416913,
2394
+ "learning_rate": 0.056050870019535494,
2395
+ "loss": 1.2546,
2396
+ "step": 341
2397
+ },
2398
+ {
2399
+ "epoch": 0.6576923076923077,
2400
+ "grad_norm": 0.004749054399683636,
2401
+ "learning_rate": 0.05549182155634076,
2402
+ "loss": 1.3886,
2403
+ "step": 342
2404
+ },
2405
+ {
2406
+ "epoch": 0.6596153846153846,
2407
+ "grad_norm": 0.004095795123847614,
2408
+ "learning_rate": 0.054934502420005464,
2409
+ "loss": 1.3486,
2410
+ "step": 343
2411
+ },
2412
+ {
2413
+ "epoch": 0.6615384615384615,
2414
+ "grad_norm": 0.004007401726655807,
2415
+ "learning_rate": 0.0543789342646837,
2416
+ "loss": 1.203,
2417
+ "step": 344
2418
+ },
2419
+ {
2420
+ "epoch": 0.6634615384615384,
2421
+ "grad_norm": 0.004329072934536201,
2422
+ "learning_rate": 0.05382513867649663,
2423
+ "loss": 1.3267,
2424
+ "step": 345
2425
+ },
2426
+ {
2427
+ "epoch": 0.6653846153846154,
2428
+ "grad_norm": 0.0038894257856872307,
2429
+ "learning_rate": 0.0532731371726938,
2430
+ "loss": 1.3364,
2431
+ "step": 346
2432
+ },
2433
+ {
2434
+ "epoch": 0.6673076923076923,
2435
+ "grad_norm": 0.00348983721037557,
2436
+ "learning_rate": 0.05272295120081732,
2437
+ "loss": 1.2275,
2438
+ "step": 347
2439
+ },
2440
+ {
2441
+ "epoch": 0.6692307692307692,
2442
+ "grad_norm": 0.004450564543363741,
2443
+ "learning_rate": 0.05217460213786821,
2444
+ "loss": 1.1814,
2445
+ "step": 348
2446
+ },
2447
+ {
2448
+ "epoch": 0.6711538461538461,
2449
+ "grad_norm": 0.004058996277773803,
2450
+ "learning_rate": 0.051628111289476024,
2451
+ "loss": 1.2309,
2452
+ "step": 349
2453
+ },
2454
+ {
2455
+ "epoch": 0.6730769230769231,
2456
+ "grad_norm": 0.003907738383479426,
2457
+ "learning_rate": 0.051083499889071106,
2458
+ "loss": 1.262,
2459
+ "step": 350
2460
+ },
2461
+ {
2462
+ "epoch": 0.675,
2463
+ "grad_norm": 0.0033577898935800783,
2464
+ "learning_rate": 0.05054078909705926,
2465
+ "loss": 1.1628,
2466
+ "step": 351
2467
+ },
2468
+ {
2469
+ "epoch": 0.676923076923077,
2470
+ "grad_norm": 0.0035955435112588574,
2471
+ "learning_rate": 0.050000000000000024,
2472
+ "loss": 1.2946,
2473
+ "step": 352
2474
+ },
2475
+ {
2476
+ "epoch": 0.6788461538461539,
2477
+ "grad_norm": 0.003133337771336547,
2478
+ "learning_rate": 0.04946115360978696,
2479
+ "loss": 1.2812,
2480
+ "step": 353
2481
+ },
2482
+ {
2483
+ "epoch": 0.6807692307692308,
2484
+ "grad_norm": 0.003772903641271153,
2485
+ "learning_rate": 0.048924270862831465,
2486
+ "loss": 1.4222,
2487
+ "step": 354
2488
+ },
2489
+ {
2490
+ "epoch": 0.6826923076923077,
2491
+ "grad_norm": 0.003690554427986993,
2492
+ "learning_rate": 0.04838937261924933,
2493
+ "loss": 1.2332,
2494
+ "step": 355
2495
+ },
2496
+ {
2497
+ "epoch": 0.6846153846153846,
2498
+ "grad_norm": 0.003783044944938042,
2499
+ "learning_rate": 0.0478564796620502,
2500
+ "loss": 1.2335,
2501
+ "step": 356
2502
+ },
2503
+ {
2504
+ "epoch": 0.6865384615384615,
2505
+ "grad_norm": 0.00322362104564831,
2506
+ "learning_rate": 0.04732561269632993,
2507
+ "loss": 1.2477,
2508
+ "step": 357
2509
+ },
2510
+ {
2511
+ "epoch": 0.6884615384615385,
2512
+ "grad_norm": 0.0037127497286017688,
2513
+ "learning_rate": 0.04679679234846636,
2514
+ "loss": 1.1825,
2515
+ "step": 358
2516
+ },
2517
+ {
2518
+ "epoch": 0.6903846153846154,
2519
+ "grad_norm": 0.004246831734802005,
2520
+ "learning_rate": 0.046270039165317606,
2521
+ "loss": 1.3536,
2522
+ "step": 359
2523
+ },
2524
+ {
2525
+ "epoch": 0.6923076923076923,
2526
+ "grad_norm": 0.004329835040577228,
2527
+ "learning_rate": 0.04574537361342407,
2528
+ "loss": 1.3676,
2529
+ "step": 360
2530
+ },
2531
+ {
2532
+ "epoch": 0.6942307692307692,
2533
+ "grad_norm": 0.003574433667834694,
2534
+ "learning_rate": 0.04522281607821288,
2535
+ "loss": 1.3535,
2536
+ "step": 361
2537
+ },
2538
+ {
2539
+ "epoch": 0.6961538461538461,
2540
+ "grad_norm": 0.0038336901083940588,
2541
+ "learning_rate": 0.04470238686320606,
2542
+ "loss": 1.2479,
2543
+ "step": 362
2544
+ },
2545
+ {
2546
+ "epoch": 0.698076923076923,
2547
+ "grad_norm": 0.0034966596078806777,
2548
+ "learning_rate": 0.044184106189231624,
2549
+ "loss": 1.2743,
2550
+ "step": 363
2551
+ },
2552
+ {
2553
+ "epoch": 0.7,
2554
+ "grad_norm": 0.0035755670646154805,
2555
+ "learning_rate": 0.043667994193637795,
2556
+ "loss": 1.3721,
2557
+ "step": 364
2558
+ },
2559
+ {
2560
+ "epoch": 0.7019230769230769,
2561
+ "grad_norm": 0.0037899478173288523,
2562
+ "learning_rate": 0.043154070929510784,
2563
+ "loss": 1.3375,
2564
+ "step": 365
2565
+ },
2566
+ {
2567
+ "epoch": 0.7038461538461539,
2568
+ "grad_norm": 0.0032568260922682902,
2569
+ "learning_rate": 0.04264235636489542,
2570
+ "loss": 1.2882,
2571
+ "step": 366
2572
+ },
2573
+ {
2574
+ "epoch": 0.7057692307692308,
2575
+ "grad_norm": 0.003532334436798711,
2576
+ "learning_rate": 0.04213287038201943,
2577
+ "loss": 1.2876,
2578
+ "step": 367
2579
+ },
2580
+ {
2581
+ "epoch": 0.7076923076923077,
2582
+ "grad_norm": 0.004250514562646025,
2583
+ "learning_rate": 0.04162563277652104,
2584
+ "loss": 1.1357,
2585
+ "step": 368
2586
+ },
2587
+ {
2588
+ "epoch": 0.7096153846153846,
2589
+ "grad_norm": 0.0034836867423431108,
2590
+ "learning_rate": 0.04112066325667954,
2591
+ "loss": 1.3264,
2592
+ "step": 369
2593
+ },
2594
+ {
2595
+ "epoch": 0.7115384615384616,
2596
+ "grad_norm": 0.003193728626562863,
2597
+ "learning_rate": 0.04061798144264986,
2598
+ "loss": 1.196,
2599
+ "step": 370
2600
+ },
2601
+ {
2602
+ "epoch": 0.7134615384615385,
2603
+ "grad_norm": 0.003650322270001297,
2604
+ "learning_rate": 0.04011760686569998,
2605
+ "loss": 1.196,
2606
+ "step": 371
2607
+ },
2608
+ {
2609
+ "epoch": 0.7153846153846154,
2610
+ "grad_norm": 0.003469423681501364,
2611
+ "learning_rate": 0.03961955896745224,
2612
+ "loss": 1.4217,
2613
+ "step": 372
2614
+ },
2615
+ {
2616
+ "epoch": 0.7173076923076923,
2617
+ "grad_norm": 0.0038995961739984315,
2618
+ "learning_rate": 0.03912385709912794,
2619
+ "loss": 1.2951,
2620
+ "step": 373
2621
+ },
2622
+ {
2623
+ "epoch": 0.7192307692307692,
2624
+ "grad_norm": 0.0035479079933222364,
2625
+ "learning_rate": 0.038630520520795276,
2626
+ "loss": 1.2786,
2627
+ "step": 374
2628
+ },
2629
+ {
2630
+ "epoch": 0.7211538461538461,
2631
+ "grad_norm": 0.0036420612751858266,
2632
+ "learning_rate": 0.03813956840062119,
2633
+ "loss": 1.1841,
2634
+ "step": 375
2635
+ },
2636
+ {
2637
+ "epoch": 0.7230769230769231,
2638
+ "grad_norm": 0.0035003379414884303,
2639
+ "learning_rate": 0.037651019814126656,
2640
+ "loss": 1.3162,
2641
+ "step": 376
2642
+ },
2643
+ {
2644
+ "epoch": 0.725,
2645
+ "grad_norm": 0.004093282239235837,
2646
+ "learning_rate": 0.037164893743445275,
2647
+ "loss": 1.2526,
2648
+ "step": 377
2649
+ },
2650
+ {
2651
+ "epoch": 0.7269230769230769,
2652
+ "grad_norm": 0.0035218341207839867,
2653
+ "learning_rate": 0.03668120907658603,
2654
+ "loss": 1.2956,
2655
+ "step": 378
2656
+ },
2657
+ {
2658
+ "epoch": 0.7288461538461538,
2659
+ "grad_norm": 0.0033165986301451553,
2660
+ "learning_rate": 0.036199984606699154,
2661
+ "loss": 1.2858,
2662
+ "step": 379
2663
+ },
2664
+ {
2665
+ "epoch": 0.7307692307692307,
2666
+ "grad_norm": 0.0045307792524292515,
2667
+ "learning_rate": 0.035721239031346066,
2668
+ "loss": 1.4073,
2669
+ "step": 380
2670
+ },
2671
+ {
2672
+ "epoch": 0.7326923076923076,
2673
+ "grad_norm": 0.003521304556723681,
2674
+ "learning_rate": 0.03524499095177297,
2675
+ "loss": 1.2761,
2676
+ "step": 381
2677
+ },
2678
+ {
2679
+ "epoch": 0.7346153846153847,
2680
+ "grad_norm": 0.003470199699451184,
2681
+ "learning_rate": 0.03477125887218792,
2682
+ "loss": 1.34,
2683
+ "step": 382
2684
+ },
2685
+ {
2686
+ "epoch": 0.7365384615384616,
2687
+ "grad_norm": 0.003914781770773064,
2688
+ "learning_rate": 0.03430006119904196,
2689
+ "loss": 1.1175,
2690
+ "step": 383
2691
+ },
2692
+ {
2693
+ "epoch": 0.7384615384615385,
2694
+ "grad_norm": 0.0037553381351364103,
2695
+ "learning_rate": 0.033831416240314084,
2696
+ "loss": 1.4548,
2697
+ "step": 384
2698
+ },
2699
+ {
2700
+ "epoch": 0.7403846153846154,
2701
+ "grad_norm": 0.0034301593182051275,
2702
+ "learning_rate": 0.03336534220479961,
2703
+ "loss": 1.2596,
2704
+ "step": 385
2705
+ },
2706
+ {
2707
+ "epoch": 0.7423076923076923,
2708
+ "grad_norm": 0.0030511838366304284,
2709
+ "learning_rate": 0.032901857201403005,
2710
+ "loss": 1.2038,
2711
+ "step": 386
2712
+ },
2713
+ {
2714
+ "epoch": 0.7442307692307693,
2715
+ "grad_norm": 0.003794444158813119,
2716
+ "learning_rate": 0.032440979238433976,
2717
+ "loss": 1.424,
2718
+ "step": 387
2719
+ },
2720
+ {
2721
+ "epoch": 0.7461538461538462,
2722
+ "grad_norm": 0.003270774595450123,
2723
+ "learning_rate": 0.03198272622290804,
2724
+ "loss": 1.1545,
2725
+ "step": 388
2726
+ },
2727
+ {
2728
+ "epoch": 0.7480769230769231,
2729
+ "grad_norm": 0.0038506840040264433,
2730
+ "learning_rate": 0.03152711595985065,
2731
+ "loss": 1.2212,
2732
+ "step": 389
2733
+ },
2734
+ {
2735
+ "epoch": 0.75,
2736
+ "grad_norm": 0.003423531138343102,
2737
+ "learning_rate": 0.031074166151605298,
2738
+ "loss": 1.2736,
2739
+ "step": 390
2740
+ },
2741
+ {
2742
+ "epoch": 0.7519230769230769,
2743
+ "grad_norm": 0.003527073480848294,
2744
+ "learning_rate": 0.030623894397145836,
2745
+ "loss": 1.3561,
2746
+ "step": 391
2747
+ },
2748
+ {
2749
+ "epoch": 0.7538461538461538,
2750
+ "grad_norm": 0.003357239266632505,
2751
+ "learning_rate": 0.03017631819139273,
2752
+ "loss": 1.1635,
2753
+ "step": 392
2754
+ },
2755
+ {
2756
+ "epoch": 0.7557692307692307,
2757
+ "grad_norm": 0.0031382644717891507,
2758
+ "learning_rate": 0.029731454924533086,
2759
+ "loss": 1.2226,
2760
+ "step": 393
2761
+ },
2762
+ {
2763
+ "epoch": 0.7576923076923077,
2764
+ "grad_norm": 0.0038842232381015107,
2765
+ "learning_rate": 0.029289321881345254,
2766
+ "loss": 1.2335,
2767
+ "step": 394
2768
+ },
2769
+ {
2770
+ "epoch": 0.7596153846153846,
2771
+ "grad_norm": 0.0038311618284447724,
2772
+ "learning_rate": 0.028849936240527008,
2773
+ "loss": 1.1898,
2774
+ "step": 395
2775
+ },
2776
+ {
2777
+ "epoch": 0.7615384615384615,
2778
+ "grad_norm": 0.0036584964032666473,
2779
+ "learning_rate": 0.028413315074028157,
2780
+ "loss": 1.2809,
2781
+ "step": 396
2782
+ },
2783
+ {
2784
+ "epoch": 0.7634615384615384,
2785
+ "grad_norm": 0.003319252047424888,
2786
+ "learning_rate": 0.027979475346387363,
2787
+ "loss": 1.2632,
2788
+ "step": 397
2789
+ },
2790
+ {
2791
+ "epoch": 0.7653846153846153,
2792
+ "grad_norm": 0.0036411846712970926,
2793
+ "learning_rate": 0.027548433914072735,
2794
+ "loss": 1.253,
2795
+ "step": 398
2796
+ },
2797
+ {
2798
+ "epoch": 0.7673076923076924,
2799
+ "grad_norm": 0.0036154057365170604,
2800
+ "learning_rate": 0.027120207524827168,
2801
+ "loss": 1.2768,
2802
+ "step": 399
2803
+ },
2804
+ {
2805
+ "epoch": 0.7692307692307693,
2806
+ "grad_norm": 0.003560065590894073,
2807
+ "learning_rate": 0.02669481281701739,
2808
+ "loss": 1.3236,
2809
+ "step": 400
2810
+ },
2811
+ {
2812
+ "epoch": 0.7711538461538462,
2813
+ "grad_norm": 0.003756163022050913,
2814
+ "learning_rate": 0.026272266318987603,
2815
+ "loss": 1.0833,
2816
+ "step": 401
2817
+ },
2818
+ {
2819
+ "epoch": 0.7730769230769231,
2820
+ "grad_norm": 0.0034826437315186158,
2821
+ "learning_rate": 0.02585258444841733,
2822
+ "loss": 1.1908,
2823
+ "step": 402
2824
+ },
2825
+ {
2826
+ "epoch": 0.775,
2827
+ "grad_norm": 0.0038305850624973064,
2828
+ "learning_rate": 0.025435783511683442,
2829
+ "loss": 1.2345,
2830
+ "step": 403
2831
+ },
2832
+ {
2833
+ "epoch": 0.7769230769230769,
2834
+ "grad_norm": 0.004439057277295419,
2835
+ "learning_rate": 0.02502187970322657,
2836
+ "loss": 1.141,
2837
+ "step": 404
2838
+ },
2839
+ {
2840
+ "epoch": 0.7788461538461539,
2841
+ "grad_norm": 0.0034870995673090158,
2842
+ "learning_rate": 0.02461088910492202,
2843
+ "loss": 1.2751,
2844
+ "step": 405
2845
+ },
2846
+ {
2847
+ "epoch": 0.7807692307692308,
2848
+ "grad_norm": 0.004207762187081535,
2849
+ "learning_rate": 0.02420282768545469,
2850
+ "loss": 1.2171,
2851
+ "step": 406
2852
+ },
2853
+ {
2854
+ "epoch": 0.7826923076923077,
2855
+ "grad_norm": 0.0035857576704798623,
2856
+ "learning_rate": 0.02379771129969892,
2857
+ "loss": 1.332,
2858
+ "step": 407
2859
+ },
2860
+ {
2861
+ "epoch": 0.7846153846153846,
2862
+ "grad_norm": 0.003778130506371736,
2863
+ "learning_rate": 0.023395555688102213,
2864
+ "loss": 1.2206,
2865
+ "step": 408
2866
+ },
2867
+ {
2868
+ "epoch": 0.7865384615384615,
2869
+ "grad_norm": 0.003977819271493016,
2870
+ "learning_rate": 0.02299637647607372,
2871
+ "loss": 1.3472,
2872
+ "step": 409
2873
+ },
2874
+ {
2875
+ "epoch": 0.7884615384615384,
2876
+ "grad_norm": 0.0033980683356138004,
2877
+ "learning_rate": 0.022600189173377264,
2878
+ "loss": 1.0608,
2879
+ "step": 410
2880
+ },
2881
+ {
2882
+ "epoch": 0.7903846153846154,
2883
+ "grad_norm": 0.0040987397350026825,
2884
+ "learning_rate": 0.022207009173528525,
2885
+ "loss": 1.3151,
2886
+ "step": 411
2887
+ },
2888
+ {
2889
+ "epoch": 0.7923076923076923,
2890
+ "grad_norm": 0.003469558217181315,
2891
+ "learning_rate": 0.02181685175319702,
2892
+ "loss": 1.2358,
2893
+ "step": 412
2894
+ },
2895
+ {
2896
+ "epoch": 0.7942307692307692,
2897
+ "grad_norm": 0.0034907360012452634,
2898
+ "learning_rate": 0.021429732071612653,
2899
+ "loss": 1.3128,
2900
+ "step": 413
2901
+ },
2902
+ {
2903
+ "epoch": 0.7961538461538461,
2904
+ "grad_norm": 0.0030503764044874015,
2905
+ "learning_rate": 0.02104566516997647,
2906
+ "loss": 1.0995,
2907
+ "step": 414
2908
+ },
2909
+ {
2910
+ "epoch": 0.7980769230769231,
2911
+ "grad_norm": 0.0034156174030708203,
2912
+ "learning_rate": 0.020664665970876496,
2913
+ "loss": 1.2098,
2914
+ "step": 415
2915
+ },
2916
+ {
2917
+ "epoch": 0.8,
2918
+ "grad_norm": 0.004078556234046615,
2919
+ "learning_rate": 0.020286749277707784,
2920
+ "loss": 1.1244,
2921
+ "step": 416
2922
+ },
2923
+ {
2924
+ "epoch": 0.801923076923077,
2925
+ "grad_norm": 0.003970691049633962,
2926
+ "learning_rate": 0.019911929774097215,
2927
+ "loss": 1.3002,
2928
+ "step": 417
2929
+ },
2930
+ {
2931
+ "epoch": 0.8038461538461539,
2932
+ "grad_norm": 0.0033691237984174627,
2933
+ "learning_rate": 0.019540222023333165,
2934
+ "loss": 1.2764,
2935
+ "step": 418
2936
+ },
2937
+ {
2938
+ "epoch": 0.8057692307692308,
2939
+ "grad_norm": 0.003921790051339545,
2940
+ "learning_rate": 0.01917164046779948,
2941
+ "loss": 1.2587,
2942
+ "step": 419
2943
+ },
2944
+ {
2945
+ "epoch": 0.8076923076923077,
2946
+ "grad_norm": 0.003665175912979948,
2947
+ "learning_rate": 0.018806199428414352,
2948
+ "loss": 1.1401,
2949
+ "step": 420
2950
+ },
2951
+ {
2952
+ "epoch": 0.8096153846153846,
2953
+ "grad_norm": 0.0036111059968629783,
2954
+ "learning_rate": 0.018443913104073985,
2955
+ "loss": 1.0759,
2956
+ "step": 421
2957
+ },
2958
+ {
2959
+ "epoch": 0.8115384615384615,
2960
+ "grad_norm": 0.004200018994974191,
2961
+ "learning_rate": 0.01808479557110081,
2962
+ "loss": 1.2039,
2963
+ "step": 422
2964
+ },
2965
+ {
2966
+ "epoch": 0.8134615384615385,
2967
+ "grad_norm": 0.004108282993784782,
2968
+ "learning_rate": 0.017728860782696667,
2969
+ "loss": 1.198,
2970
+ "step": 423
2971
+ },
2972
+ {
2973
+ "epoch": 0.8153846153846154,
2974
+ "grad_norm": 0.0041163982563187295,
2975
+ "learning_rate": 0.017376122568400532,
2976
+ "loss": 1.39,
2977
+ "step": 424
2978
+ },
2979
+ {
2980
+ "epoch": 0.8173076923076923,
2981
+ "grad_norm": 0.0033421297793850544,
2982
+ "learning_rate": 0.017026594633551252,
2983
+ "loss": 1.1977,
2984
+ "step": 425
2985
+ },
2986
+ {
2987
+ "epoch": 0.8192307692307692,
2988
+ "grad_norm": 0.0047570189827926285,
2989
+ "learning_rate": 0.01668029055875512,
2990
+ "loss": 1.2293,
2991
+ "step": 426
2992
+ },
2993
+ {
2994
+ "epoch": 0.8211538461538461,
2995
+ "grad_norm": 0.0035298297268954813,
2996
+ "learning_rate": 0.016337223799358026,
2997
+ "loss": 1.1343,
2998
+ "step": 427
2999
+ },
3000
+ {
3001
+ "epoch": 0.823076923076923,
3002
+ "grad_norm": 0.0035230260272556824,
3003
+ "learning_rate": 0.01599740768492286,
3004
+ "loss": 1.1063,
3005
+ "step": 428
3006
+ },
3007
+ {
3008
+ "epoch": 0.825,
3009
+ "grad_norm": 0.0034156130687512565,
3010
+ "learning_rate": 0.015660855418711452,
3011
+ "loss": 1.2172,
3012
+ "step": 429
3013
+ },
3014
+ {
3015
+ "epoch": 0.8269230769230769,
3016
+ "grad_norm": 0.0031794976829662243,
3017
+ "learning_rate": 0.015327580077171589,
3018
+ "loss": 1.2042,
3019
+ "step": 430
3020
+ },
3021
+ {
3022
+ "epoch": 0.8288461538461539,
3023
+ "grad_norm": 0.003986666060483522,
3024
+ "learning_rate": 0.014997594609429088,
3025
+ "loss": 1.2684,
3026
+ "step": 431
3027
+ },
3028
+ {
3029
+ "epoch": 0.8307692307692308,
3030
+ "grad_norm": 0.003883959511710326,
3031
+ "learning_rate": 0.01467091183678444,
3032
+ "loss": 1.1139,
3033
+ "step": 432
3034
+ },
3035
+ {
3036
+ "epoch": 0.8326923076923077,
3037
+ "grad_norm": 0.0035282434565686175,
3038
+ "learning_rate": 0.014347544452214867,
3039
+ "loss": 1.2569,
3040
+ "step": 433
3041
+ },
3042
+ {
3043
+ "epoch": 0.8346153846153846,
3044
+ "grad_norm": 0.003460904613025027,
3045
+ "learning_rate": 0.014027505019880971,
3046
+ "loss": 0.9965,
3047
+ "step": 434
3048
+ },
3049
+ {
3050
+ "epoch": 0.8365384615384616,
3051
+ "grad_norm": 0.0041950218722731,
3052
+ "learning_rate": 0.013710805974638696,
3053
+ "loss": 1.2884,
3054
+ "step": 435
3055
+ },
3056
+ {
3057
+ "epoch": 0.8384615384615385,
3058
+ "grad_norm": 0.0034138050890129227,
3059
+ "learning_rate": 0.01339745962155613,
3060
+ "loss": 1.0811,
3061
+ "step": 436
3062
+ },
3063
+ {
3064
+ "epoch": 0.8403846153846154,
3065
+ "grad_norm": 0.003510781950412414,
3066
+ "learning_rate": 0.01308747813543536,
3067
+ "loss": 1.317,
3068
+ "step": 437
3069
+ },
3070
+ {
3071
+ "epoch": 0.8423076923076923,
3072
+ "grad_norm": 0.003504194003221874,
3073
+ "learning_rate": 0.012780873560339467,
3074
+ "loss": 1.1184,
3075
+ "step": 438
3076
+ },
3077
+ {
3078
+ "epoch": 0.8442307692307692,
3079
+ "grad_norm": 0.0036570760391396785,
3080
+ "learning_rate": 0.012477657809124632,
3081
+ "loss": 1.2422,
3082
+ "step": 439
3083
+ },
3084
+ {
3085
+ "epoch": 0.8461538461538461,
3086
+ "grad_norm": 0.0036298077381627936,
3087
+ "learning_rate": 0.012177842662977134,
3088
+ "loss": 1.1734,
3089
+ "step": 440
3090
+ },
3091
+ {
3092
+ "epoch": 0.8480769230769231,
3093
+ "grad_norm": 0.003427738039683916,
3094
+ "learning_rate": 0.01188143977095576,
3095
+ "loss": 1.2688,
3096
+ "step": 441
3097
+ },
3098
+ {
3099
+ "epoch": 0.85,
3100
+ "grad_norm": 0.004035363684542472,
3101
+ "learning_rate": 0.011588460649539035,
3102
+ "loss": 1.2219,
3103
+ "step": 442
3104
+ },
3105
+ {
3106
+ "epoch": 0.8519230769230769,
3107
+ "grad_norm": 0.0036497185731645863,
3108
+ "learning_rate": 0.011298916682177829,
3109
+ "loss": 1.2482,
3110
+ "step": 443
3111
+ },
3112
+ {
3113
+ "epoch": 0.8538461538461538,
3114
+ "grad_norm": 0.0032078940141462057,
3115
+ "learning_rate": 0.011012819118853146,
3116
+ "loss": 1.2124,
3117
+ "step": 444
3118
+ },
3119
+ {
3120
+ "epoch": 0.8557692307692307,
3121
+ "grad_norm": 0.0037970210525749156,
3122
+ "learning_rate": 0.01073017907563887,
3123
+ "loss": 1.1332,
3124
+ "step": 445
3125
+ },
3126
+ {
3127
+ "epoch": 0.8576923076923076,
3128
+ "grad_norm": 0.003631064068066613,
3129
+ "learning_rate": 0.010451007534269908,
3130
+ "loss": 1.347,
3131
+ "step": 446
3132
+ },
3133
+ {
3134
+ "epoch": 0.8596153846153847,
3135
+ "grad_norm": 0.0035187214113329775,
3136
+ "learning_rate": 0.010175315341715598,
3137
+ "loss": 1.2208,
3138
+ "step": 447
3139
+ },
3140
+ {
3141
+ "epoch": 0.8615384615384616,
3142
+ "grad_norm": 0.0037696600541564907,
3143
+ "learning_rate": 0.009903113209758098,
3144
+ "loss": 1.1975,
3145
+ "step": 448
3146
+ },
3147
+ {
3148
+ "epoch": 0.8634615384615385,
3149
+ "grad_norm": 0.0038683075228141558,
3150
+ "learning_rate": 0.009634411714576352,
3151
+ "loss": 1.3012,
3152
+ "step": 449
3153
+ },
3154
+ {
3155
+ "epoch": 0.8653846153846154,
3156
+ "grad_norm": 0.003524874636975072,
3157
+ "learning_rate": 0.009369221296335007,
3158
+ "loss": 1.2395,
3159
+ "step": 450
3160
+ },
3161
+ {
3162
+ "epoch": 0.8673076923076923,
3163
+ "grad_norm": 0.0035359024345832505,
3164
+ "learning_rate": 0.009107552258778906,
3165
+ "loss": 1.2331,
3166
+ "step": 451
3167
+ },
3168
+ {
3169
+ "epoch": 0.8692307692307693,
3170
+ "grad_norm": 0.0034982654113694927,
3171
+ "learning_rate": 0.008849414768832687,
3172
+ "loss": 1.3385,
3173
+ "step": 452
3174
+ },
3175
+ {
3176
+ "epoch": 0.8711538461538462,
3177
+ "grad_norm": 0.00409030519497401,
3178
+ "learning_rate": 0.008594818856205699,
3179
+ "loss": 1.3142,
3180
+ "step": 453
3181
+ },
3182
+ {
3183
+ "epoch": 0.8730769230769231,
3184
+ "grad_norm": 0.003987596593164173,
3185
+ "learning_rate": 0.00834377441300238,
3186
+ "loss": 1.1473,
3187
+ "step": 454
3188
+ },
3189
+ {
3190
+ "epoch": 0.875,
3191
+ "grad_norm": 0.003500946909265335,
3192
+ "learning_rate": 0.008096291193337934,
3193
+ "loss": 1.2781,
3194
+ "step": 455
3195
+ },
3196
+ {
3197
+ "epoch": 0.8769230769230769,
3198
+ "grad_norm": 0.003532610444754912,
3199
+ "learning_rate": 0.007852378812959226,
3200
+ "loss": 1.1993,
3201
+ "step": 456
3202
+ },
3203
+ {
3204
+ "epoch": 0.8788461538461538,
3205
+ "grad_norm": 0.0032370576830697893,
3206
+ "learning_rate": 0.007612046748871327,
3207
+ "loss": 1.2843,
3208
+ "step": 457
3209
+ },
3210
+ {
3211
+ "epoch": 0.8807692307692307,
3212
+ "grad_norm": 0.0036384162319098878,
3213
+ "learning_rate": 0.007375304338969136,
3214
+ "loss": 1.3413,
3215
+ "step": 458
3216
+ },
3217
+ {
3218
+ "epoch": 0.8826923076923077,
3219
+ "grad_norm": 0.003613724445135647,
3220
+ "learning_rate": 0.007142160781674645,
3221
+ "loss": 1.2729,
3222
+ "step": 459
3223
+ },
3224
+ {
3225
+ "epoch": 0.8846153846153846,
3226
+ "grad_norm": 0.003345630142979864,
3227
+ "learning_rate": 0.006912625135579587,
3228
+ "loss": 1.1419,
3229
+ "step": 460
3230
+ },
3231
+ {
3232
+ "epoch": 0.8865384615384615,
3233
+ "grad_norm": 0.002938815382097615,
3234
+ "learning_rate": 0.0066867063190933496,
3235
+ "loss": 1.3631,
3236
+ "step": 461
3237
+ },
3238
+ {
3239
+ "epoch": 0.8884615384615384,
3240
+ "grad_norm": 0.0035575626293444214,
3241
+ "learning_rate": 0.006464413110096601,
3242
+ "loss": 1.3936,
3243
+ "step": 462
3244
+ },
3245
+ {
3246
+ "epoch": 0.8903846153846153,
3247
+ "grad_norm": 0.00371955264900499,
3248
+ "learning_rate": 0.006245754145600091,
3249
+ "loss": 1.1077,
3250
+ "step": 463
3251
+ },
3252
+ {
3253
+ "epoch": 0.8923076923076924,
3254
+ "grad_norm": 0.003969204728630893,
3255
+ "learning_rate": 0.006030737921409169,
3256
+ "loss": 1.2538,
3257
+ "step": 464
3258
+ },
3259
+ {
3260
+ "epoch": 0.8942307692307693,
3261
+ "grad_norm": 0.004209766463495701,
3262
+ "learning_rate": 0.005819372791793654,
3263
+ "loss": 1.3737,
3264
+ "step": 465
3265
+ },
3266
+ {
3267
+ "epoch": 0.8961538461538462,
3268
+ "grad_norm": 0.0031720047591087632,
3269
+ "learning_rate": 0.005611666969163243,
3270
+ "loss": 1.2384,
3271
+ "step": 466
3272
+ },
3273
+ {
3274
+ "epoch": 0.8980769230769231,
3275
+ "grad_norm": 0.0035474650563144723,
3276
+ "learning_rate": 0.005407628523748398,
3277
+ "loss": 1.2688,
3278
+ "step": 467
3279
+ },
3280
+ {
3281
+ "epoch": 0.9,
3282
+ "grad_norm": 0.0039817084422281886,
3283
+ "learning_rate": 0.00520726538328683,
3284
+ "loss": 1.2168,
3285
+ "step": 468
3286
+ },
3287
+ {
3288
+ "epoch": 0.9019230769230769,
3289
+ "grad_norm": 0.003696546865726876,
3290
+ "learning_rate": 0.005010585332715401,
3291
+ "loss": 1.2736,
3292
+ "step": 469
3293
+ },
3294
+ {
3295
+ "epoch": 0.9038461538461539,
3296
+ "grad_norm": 0.00310681148113461,
3297
+ "learning_rate": 0.004817596013867765,
3298
+ "loss": 1.1531,
3299
+ "step": 470
3300
+ },
3301
+ {
3302
+ "epoch": 0.9057692307692308,
3303
+ "grad_norm": 0.003808577545178428,
3304
+ "learning_rate": 0.004628304925177318,
3305
+ "loss": 1.1657,
3306
+ "step": 471
3307
+ },
3308
+ {
3309
+ "epoch": 0.9076923076923077,
3310
+ "grad_norm": 0.0035338411317336824,
3311
+ "learning_rate": 0.004442719421385921,
3312
+ "loss": 1.1405,
3313
+ "step": 472
3314
+ },
3315
+ {
3316
+ "epoch": 0.9096153846153846,
3317
+ "grad_norm": 0.003523657496870884,
3318
+ "learning_rate": 0.004260846713258193,
3319
+ "loss": 1.2013,
3320
+ "step": 473
3321
+ },
3322
+ {
3323
+ "epoch": 0.9115384615384615,
3324
+ "grad_norm": 0.003559754796752954,
3325
+ "learning_rate": 0.004082693867301224,
3326
+ "loss": 1.304,
3327
+ "step": 474
3328
+ },
3329
+ {
3330
+ "epoch": 0.9134615384615384,
3331
+ "grad_norm": 0.003325501476995395,
3332
+ "learning_rate": 0.003908267805490051,
3333
+ "loss": 1.2181,
3334
+ "step": 475
3335
+ },
3336
+ {
3337
+ "epoch": 0.9153846153846154,
3338
+ "grad_norm": 0.0035885172855855347,
3339
+ "learning_rate": 0.003737575304998797,
3340
+ "loss": 1.1954,
3341
+ "step": 476
3342
+ },
3343
+ {
3344
+ "epoch": 0.9173076923076923,
3345
+ "grad_norm": 0.004089913587604459,
3346
+ "learning_rate": 0.003570622997937234,
3347
+ "loss": 1.1842,
3348
+ "step": 477
3349
+ },
3350
+ {
3351
+ "epoch": 0.9192307692307692,
3352
+ "grad_norm": 0.0034228267975752492,
3353
+ "learning_rate": 0.00340741737109318,
3354
+ "loss": 1.1438,
3355
+ "step": 478
3356
+ },
3357
+ {
3358
+ "epoch": 0.9211538461538461,
3359
+ "grad_norm": 0.0037384365581522934,
3360
+ "learning_rate": 0.003247964765680389,
3361
+ "loss": 1.272,
3362
+ "step": 479
3363
+ },
3364
+ {
3365
+ "epoch": 0.9230769230769231,
3366
+ "grad_norm": 0.003430574548015608,
3367
+ "learning_rate": 0.0030922713770922153,
3368
+ "loss": 1.2936,
3369
+ "step": 480
3370
+ },
3371
+ {
3372
+ "epoch": 0.925,
3373
+ "grad_norm": 0.00415151088301985,
3374
+ "learning_rate": 0.0029403432546609046,
3375
+ "loss": 1.3122,
3376
+ "step": 481
3377
+ },
3378
+ {
3379
+ "epoch": 0.926923076923077,
3380
+ "grad_norm": 0.003568633083861599,
3381
+ "learning_rate": 0.0027921863014225504,
3382
+ "loss": 1.3083,
3383
+ "step": 482
3384
+ },
3385
+ {
3386
+ "epoch": 0.9288461538461539,
3387
+ "grad_norm": 0.003658033164292622,
3388
+ "learning_rate": 0.002647806273887665,
3389
+ "loss": 1.2159,
3390
+ "step": 483
3391
+ },
3392
+ {
3393
+ "epoch": 0.9307692307692308,
3394
+ "grad_norm": 0.003525495510625043,
3395
+ "learning_rate": 0.0025072087818176383,
3396
+ "loss": 1.2171,
3397
+ "step": 484
3398
+ },
3399
+ {
3400
+ "epoch": 0.9326923076923077,
3401
+ "grad_norm": 0.003317192493476743,
3402
+ "learning_rate": 0.002370399288006664,
3403
+ "loss": 1.1647,
3404
+ "step": 485
3405
+ },
3406
+ {
3407
+ "epoch": 0.9346153846153846,
3408
+ "grad_norm": 0.0037105624503438583,
3409
+ "learning_rate": 0.0022373831080695463,
3410
+ "loss": 1.2858,
3411
+ "step": 486
3412
+ },
3413
+ {
3414
+ "epoch": 0.9365384615384615,
3415
+ "grad_norm": 0.003356454130411601,
3416
+ "learning_rate": 0.0021081654102351635,
3417
+ "loss": 1.1415,
3418
+ "step": 487
3419
+ },
3420
+ {
3421
+ "epoch": 0.9384615384615385,
3422
+ "grad_norm": 0.003553510582982068,
3423
+ "learning_rate": 0.0019827512151456175,
3424
+ "loss": 1.0784,
3425
+ "step": 488
3426
+ },
3427
+ {
3428
+ "epoch": 0.9403846153846154,
3429
+ "grad_norm": 0.0031229299804875076,
3430
+ "learning_rate": 0.0018611453956612345,
3431
+ "loss": 1.2956,
3432
+ "step": 489
3433
+ },
3434
+ {
3435
+ "epoch": 0.9423076923076923,
3436
+ "grad_norm": 0.0037783687269465557,
3437
+ "learning_rate": 0.0017433526766711727,
3438
+ "loss": 1.2088,
3439
+ "step": 490
3440
+ },
3441
+ {
3442
+ "epoch": 0.9442307692307692,
3443
+ "grad_norm": 0.003541990543434244,
3444
+ "learning_rate": 0.0016293776349098677,
3445
+ "loss": 1.1657,
3446
+ "step": 491
3447
+ },
3448
+ {
3449
+ "epoch": 0.9461538461538461,
3450
+ "grad_norm": 0.0036870960587380612,
3451
+ "learning_rate": 0.0015192246987791981,
3452
+ "loss": 1.2896,
3453
+ "step": 492
3454
+ },
3455
+ {
3456
+ "epoch": 0.948076923076923,
3457
+ "grad_norm": 0.003790922597235061,
3458
+ "learning_rate": 0.0014128981481764114,
3459
+ "loss": 1.3428,
3460
+ "step": 493
3461
+ },
3462
+ {
3463
+ "epoch": 0.95,
3464
+ "grad_norm": 0.0031990804938886157,
3465
+ "learning_rate": 0.0013104021143278911,
3466
+ "loss": 1.2225,
3467
+ "step": 494
3468
+ },
3469
+ {
3470
+ "epoch": 0.9519230769230769,
3471
+ "grad_norm": 0.003378492635097867,
3472
+ "learning_rate": 0.0012117405796285285,
3473
+ "loss": 1.1823,
3474
+ "step": 495
3475
+ },
3476
+ {
3477
+ "epoch": 0.9538461538461539,
3478
+ "grad_norm": 0.0037970363368346675,
3479
+ "learning_rate": 0.0011169173774871477,
3480
+ "loss": 1.1005,
3481
+ "step": 496
3482
+ },
3483
+ {
3484
+ "epoch": 0.9557692307692308,
3485
+ "grad_norm": 0.0030747185553710804,
3486
+ "learning_rate": 0.0010259361921774012,
3487
+ "loss": 1.2266,
3488
+ "step": 497
3489
+ },
3490
+ {
3491
+ "epoch": 0.9576923076923077,
3492
+ "grad_norm": 0.003618738277238636,
3493
+ "learning_rate": 0.000938800558694719,
3494
+ "loss": 1.1842,
3495
+ "step": 498
3496
+ },
3497
+ {
3498
+ "epoch": 0.9596153846153846,
3499
+ "grad_norm": 0.0033870563573956,
3500
+ "learning_rate": 0.0008555138626189618,
3501
+ "loss": 1.3705,
3502
+ "step": 499
3503
+ },
3504
+ {
3505
+ "epoch": 0.9615384615384616,
3506
+ "grad_norm": 0.00404793147494844,
3507
+ "learning_rate": 0.0007760793399827937,
3508
+ "loss": 1.3031,
3509
+ "step": 500
3510
+ },
3511
+ {
3512
+ "epoch": 0.9634615384615385,
3513
+ "grad_norm": 0.004515974499268608,
3514
+ "learning_rate": 0.000700500077146038,
3515
+ "loss": 1.2834,
3516
+ "step": 501
3517
+ },
3518
+ {
3519
+ "epoch": 0.9653846153846154,
3520
+ "grad_norm": 0.0033328077649257375,
3521
+ "learning_rate": 0.0006287790106757397,
3522
+ "loss": 1.2227,
3523
+ "step": 502
3524
+ },
3525
+ {
3526
+ "epoch": 0.9673076923076923,
3527
+ "grad_norm": 0.003548638344190246,
3528
+ "learning_rate": 0.0005609189272320237,
3529
+ "loss": 1.2607,
3530
+ "step": 503
3531
+ },
3532
+ {
3533
+ "epoch": 0.9692307692307692,
3534
+ "grad_norm": 0.00411562561829213,
3535
+ "learning_rate": 0.000496922463459859,
3536
+ "loss": 1.2188,
3537
+ "step": 504
3538
+ },
3539
+ {
3540
+ "epoch": 0.9711538461538461,
3541
+ "grad_norm": 0.0034516307884759926,
3542
+ "learning_rate": 0.0004367921058866187,
3543
+ "loss": 1.2604,
3544
+ "step": 505
3545
+ },
3546
+ {
3547
+ "epoch": 0.9730769230769231,
3548
+ "grad_norm": 0.0038699819333364864,
3549
+ "learning_rate": 0.0003805301908254455,
3550
+ "loss": 1.1757,
3551
+ "step": 506
3552
+ },
3553
+ {
3554
+ "epoch": 0.975,
3555
+ "grad_norm": 0.0033297748654323974,
3556
+ "learning_rate": 0.0003281389042844918,
3557
+ "loss": 1.4127,
3558
+ "step": 507
3559
+ },
3560
+ {
3561
+ "epoch": 0.9769230769230769,
3562
+ "grad_norm": 0.003502188296335559,
3563
+ "learning_rate": 0.00027962028188198705,
3564
+ "loss": 1.2916,
3565
+ "step": 508
3566
+ },
3567
+ {
3568
+ "epoch": 0.9788461538461538,
3569
+ "grad_norm": 0.0035264476471048123,
3570
+ "learning_rate": 0.00023497620876711256,
3571
+ "loss": 1.262,
3572
+ "step": 509
3573
+ },
3574
+ {
3575
+ "epoch": 0.9807692307692307,
3576
+ "grad_norm": 0.003438073221406764,
3577
+ "learning_rate": 0.00019420841954681523,
3578
+ "loss": 1.2168,
3579
+ "step": 510
3580
+ },
3581
+ {
3582
+ "epoch": 0.9826923076923076,
3583
+ "grad_norm": 0.0033738006855417303,
3584
+ "learning_rate": 0.00015731849821833956,
3585
+ "loss": 1.191,
3586
+ "step": 511
3587
+ },
3588
+ {
3589
+ "epoch": 0.9846153846153847,
3590
+ "grad_norm": 0.0033485634222802426,
3591
+ "learning_rate": 0.00012430787810776555,
3592
+ "loss": 1.0719,
3593
+ "step": 512
3594
+ },
3595
+ {
3596
+ "epoch": 0.9865384615384616,
3597
+ "grad_norm": 0.003844301618000297,
3598
+ "learning_rate": 9.517784181422018e-05,
3599
+ "loss": 1.2758,
3600
+ "step": 513
3601
+ },
3602
+ {
3603
+ "epoch": 0.9884615384615385,
3604
+ "grad_norm": 0.003187262147242988,
3605
+ "learning_rate": 6.992952116013917e-05,
3606
+ "loss": 1.2525,
3607
+ "step": 514
3608
+ },
3609
+ {
3610
+ "epoch": 0.9903846153846154,
3611
+ "grad_norm": 0.00409744429691666,
3612
+ "learning_rate": 4.856389714723575e-05,
3613
+ "loss": 1.2985,
3614
+ "step": 515
3615
+ },
3616
+ {
3617
+ "epoch": 0.9923076923076923,
3618
+ "grad_norm": 0.0033934119238889903,
3619
+ "learning_rate": 3.108179991837545e-05,
3620
+ "loss": 1.1857,
3621
+ "step": 516
3622
+ },
3623
+ {
3624
+ "epoch": 0.9942307692307693,
3625
+ "grad_norm": 0.004120843756187969,
3626
+ "learning_rate": 1.7483908725357544e-05,
3627
+ "loss": 1.327,
3628
+ "step": 517
3629
+ },
3630
+ {
3631
+ "epoch": 0.9961538461538462,
3632
+ "grad_norm": 0.003707080248010137,
3633
+ "learning_rate": 7.770751902513862e-06,
3634
+ "loss": 1.2072,
3635
+ "step": 518
3636
+ },
3637
+ {
3638
+ "epoch": 0.9980769230769231,
3639
+ "grad_norm": 0.0036586293830800313,
3640
+ "learning_rate": 1.9427068461808086e-06,
3641
+ "loss": 1.2618,
3642
+ "step": 519
3643
+ },
3644
+ {
3645
+ "epoch": 1.0,
3646
+ "grad_norm": 0.0038020248592507125,
3647
+ "learning_rate": 0.0,
3648
+ "loss": 1.3377,
3649
+ "step": 520
3650
+ },
3651
+ {
3652
+ "epoch": 1.0,
3653
+ "step": 520,
3654
+ "total_flos": 419724754747392.0,
3655
+ "train_loss": 1.4914764440976656,
3656
+ "train_runtime": 1938.4042,
3657
+ "train_samples_per_second": 34.322,
3658
+ "train_steps_per_second": 0.268
3659
+ }
3660
+ ],
3661
+ "logging_steps": 1.0,
3662
+ "max_steps": 520,
3663
+ "num_input_tokens_seen": 0,
3664
+ "num_train_epochs": 1,
3665
+ "save_steps": 50000,
3666
+ "total_flos": 419724754747392.0,
3667
+ "train_batch_size": 4,
3668
+ "trial_name": null,
3669
+ "trial_params": null
3670
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d321ca2300fe53f4e9bb79ca1dd87c134829a2189348a46b87a809bfdb8f920
3
+ size 7608
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.3_2e-1_connector-1.0_2.3_2e-1_ablation/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/config.json ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TinyLlavaForConditionalGeneration"
4
+ ],
5
+ "backward_type_connector": "normal",
6
+ "cache_dir": null,
7
+ "connector_type": "mlp2x_gelu",
8
+ "hidden_size": 896,
9
+ "ignore_index": -100,
10
+ "image_aspect_ratio": "square",
11
+ "image_token_index": -200,
12
+ "llm_model_name_or_path": "Qwen/Qwen2.5-0.5B",
13
+ "mask_model": [
14
+ "llm",
15
+ "connector"
16
+ ],
17
+ "mask_type_connector": "soft",
18
+ "model_type": "tinyllava",
19
+ "num_queries": 128,
20
+ "num_resampler_layers": 3,
21
+ "pad_token": "<|endoftext|>",
22
+ "pad_token_id": 151643,
23
+ "resampler_hidden_size": 768,
24
+ "sparsity_connector": null,
25
+ "subnet_type_connector": "global",
26
+ "temperature_connector": 2.9,
27
+ "text_config": {
28
+ "_name_or_path": "Qwen/Qwen2.5-0.5B",
29
+ "architectures": [
30
+ "Qwen2ForCausalLM"
31
+ ],
32
+ "backward_type": "normal",
33
+ "bos_token_id": 151643,
34
+ "eos_token_id": 151643,
35
+ "hidden_size": 896,
36
+ "intermediate_size": 4864,
37
+ "mask_type": "soft",
38
+ "masked_layers": "all",
39
+ "max_position_embeddings": 32768,
40
+ "max_window_layers": 24,
41
+ "model_type": "qwen2",
42
+ "num_attention_heads": 14,
43
+ "num_hidden_layers": 24,
44
+ "num_key_value_heads": 2,
45
+ "rope_theta": 1000000.0,
46
+ "sliding_window": 32768,
47
+ "subnet_mode": "both",
48
+ "subnet_type": "None",
49
+ "temperature_attn": 2.9,
50
+ "temperature_mlp": 2.9,
51
+ "tie_word_embeddings": true,
52
+ "torch_dtype": "bfloat16",
53
+ "use_mrope": false,
54
+ "use_sliding_window": false,
55
+ "vocab_size": 151936
56
+ },
57
+ "threshold_connector": null,
58
+ "tokenizer_model_max_length": 2048,
59
+ "tokenizer_name_or_path": "Qwen/Qwen2.5-0.5B",
60
+ "tokenizer_padding_side": "right",
61
+ "tokenizer_use_fast": false,
62
+ "torch_dtype": "bfloat16",
63
+ "transformers_version": "4.40.1",
64
+ "tune_type_connector": "full",
65
+ "tune_type_llm": "full",
66
+ "tune_type_vision_tower": "frozen",
67
+ "tune_vision_tower_from_layer": 0,
68
+ "use_cache": true,
69
+ "vision_config": {
70
+ "hidden_act": "gelu_pytorch_tanh",
71
+ "hidden_size": 1152,
72
+ "image_size": 384,
73
+ "intermediate_size": 4304,
74
+ "layer_norm_eps": 1e-06,
75
+ "model_name_or_path": "google/siglip-so400m-patch14-384",
76
+ "model_name_or_path2": "",
77
+ "model_type": "siglip_vision_model",
78
+ "num_attention_heads": 16,
79
+ "num_hidden_layers": 27,
80
+ "patch_size": 14
81
+ },
82
+ "vision_feature_layer": -2,
83
+ "vision_feature_select_strategy": "patch",
84
+ "vision_hidden_size": 1152,
85
+ "vision_model_name_or_path": "google/siglip-so400m-patch14-384",
86
+ "vision_model_name_or_path2": "",
87
+ "vocab_size": 151936
88
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151643,
4
+ "eos_token_id": 151643,
5
+ "transformers_version": "4.40.1",
6
+ "use_cache": false
7
+ }
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/log.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ 2025-10-17 09:41:41,800 | INFO: Total Parameters: 1283756736, Total Trainable Parameters: 359661568
2
+ 2025-10-17 09:41:41,809 | INFO: Trainable Parameters:
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:380a1a71b18445f69d0b0d263af22f9d3a63d54872ec27bd78fff4a46d4b0548
3
+ size 2567640568
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/runs/Oct17_09-25-10_ywang29-vrdb-test1-worker-0/events.out.tfevents.1760694106.ywang29-vrdb-test1-worker-0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2b4d5baaca21224c9d53b48ec00f7da13078cf76417805056850992a286e9f6
3
+ size 116893
ckpts_oct17/qwen2.5-0_5b_base_masktune_42_llm-connector_text-1.0_2.9_2e-1_connector-1.0_2.9_2e-1_ablation/special_tokens_map.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|endoftext|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ },
31
+ "unk_token": "<|endoftext|>"
32
+ }