llama2_instruct_generation
This model is a fine-tuned version of NousResearch/Llama-2-7b-hf on the generator dataset.
It achieves the following results on the evaluation set:
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 20
- training_steps: 500
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
| 1.9453 |
0.0 |
20 |
1.8075 |
| 1.8785 |
0.01 |
40 |
1.7784 |
| 1.8857 |
0.01 |
60 |
1.7662 |
| 1.8799 |
0.01 |
80 |
1.7530 |
| 1.9231 |
0.01 |
100 |
1.7425 |
| 1.813 |
0.02 |
120 |
1.7142 |
| 1.8755 |
0.02 |
140 |
1.7035 |
| 1.8051 |
0.02 |
160 |
1.6975 |
| 1.7347 |
0.02 |
180 |
1.6927 |
| 1.7275 |
0.03 |
200 |
1.6907 |
| 1.8213 |
0.03 |
220 |
1.6882 |
| 1.7671 |
0.03 |
240 |
1.6851 |
| 1.7736 |
0.04 |
260 |
1.6839 |
| 1.8085 |
0.04 |
280 |
1.6826 |
| 1.74 |
0.04 |
300 |
1.6803 |
| 1.7354 |
0.04 |
320 |
1.6789 |
| 1.7736 |
0.05 |
340 |
1.6808 |
| 1.8226 |
0.05 |
360 |
1.6786 |
| 1.7954 |
0.05 |
380 |
1.6761 |
| 1.7472 |
0.05 |
400 |
1.6746 |
| 1.6613 |
0.06 |
420 |
1.6736 |
| 1.7523 |
0.06 |
440 |
1.6721 |
| 1.7296 |
0.06 |
460 |
1.6717 |
| 1.8324 |
0.07 |
480 |
1.6729 |
| 1.7083 |
0.07 |
500 |
1.6717 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
