| --- |
| library_name: peft |
| tags: |
| - trl |
| - sft |
| - generated_from_trainer |
| datasets: |
| - generator |
| base_model: NousResearch/Llama-2-7b-hf |
| model-index: |
| - name: llama2_instruct_generation |
| results: [] |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| # llama2_instruct_generation |
|
|
| This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on the generator dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 1.6728 |
|
|
| ## 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: 0.03 |
| - training_steps: 500 |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | |
| |:-------------:|:-----:|:----:|:---------------:| |
| | 1.9239 | 0.0 | 20 | 1.8081 | |
| | 1.8915 | 0.01 | 40 | 1.7801 | |
| | 1.9622 | 0.01 | 60 | 1.7659 | |
| | 1.8338 | 0.01 | 80 | 1.7555 | |
| | 1.8614 | 0.01 | 100 | 1.7390 | |
| | 1.8221 | 0.02 | 120 | 1.7068 | |
| | 1.7601 | 0.02 | 140 | 1.7031 | |
| | 1.8557 | 0.02 | 160 | 1.6985 | |
| | 1.8575 | 0.02 | 180 | 1.6942 | |
| | 1.777 | 0.03 | 200 | 1.6925 | |
| | 1.8087 | 0.03 | 220 | 1.6904 | |
| | 1.856 | 0.03 | 240 | 1.6884 | |
| | 1.7704 | 0.04 | 260 | 1.6870 | |
| | 1.819 | 0.04 | 280 | 1.6838 | |
| | 1.8136 | 0.04 | 300 | 1.6836 | |
| | 1.768 | 0.04 | 320 | 1.6821 | |
| | 1.7937 | 0.05 | 340 | 1.6809 | |
| | 1.8045 | 0.05 | 360 | 1.6791 | |
| | 1.7958 | 0.05 | 380 | 1.6784 | |
| | 1.7995 | 0.05 | 400 | 1.6790 | |
| | 1.805 | 0.06 | 420 | 1.6775 | |
| | 1.9388 | 0.06 | 440 | 1.6742 | |
| | 1.8304 | 0.06 | 460 | 1.6742 | |
| | 1.8732 | 0.07 | 480 | 1.6729 | |
| | 1.8443 | 0.07 | 500 | 1.6728 | |
| |
| |
| ### Framework versions |
| |
| - PEFT 0.8.2 |
| - Transformers 4.37.2 |
| - Pytorch 2.1.0+cu121 |
| - Datasets 2.16.1 |
| - Tokenizers 0.15.1 |
