| | --- |
| | license: other |
| | base_model: yahma/llama-7b-hf |
| | tags: |
| | - generated_from_trainer |
| | model-index: |
| | - name: V0224P8 |
| | 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. --> |
| |
|
| | # V0224P8 |
| |
|
| | This model is a fine-tuned version of [yahma/llama-7b-hf](https://huggingface.co/yahma/llama-7b-hf) on an unknown dataset. |
| | It achieves the following results on the evaluation set: |
| | - Loss: 0.7483 |
| |
|
| | ## 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: 0.0003 |
| | - train_batch_size: 4 |
| | - eval_batch_size: 8 |
| | - seed: 42 |
| | - gradient_accumulation_steps: 32 |
| | - total_train_batch_size: 128 |
| | - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| | - lr_scheduler_type: cosine_with_restarts |
| | - lr_scheduler_warmup_steps: 20 |
| | - num_epochs: 3 |
| | - mixed_precision_training: Native AMP |
| |
|
| | ### Training results |
| |
|
| | | Training Loss | Epoch | Step | Validation Loss | |
| | |:-------------:|:-----:|:----:|:---------------:| |
| | | 1.0988 | 0.13 | 10 | 1.0588 | |
| | | 0.9892 | 0.26 | 20 | 0.9511 | |
| | | 0.9008 | 0.39 | 30 | 0.8673 | |
| | | 0.8366 | 0.52 | 40 | 0.8299 | |
| | | 0.8139 | 0.65 | 50 | 0.8099 | |
| | | 0.7954 | 0.78 | 60 | 0.7970 | |
| | | 0.7704 | 0.91 | 70 | 0.7862 | |
| | | 0.7614 | 1.04 | 80 | 0.7788 | |
| | | 0.7322 | 1.17 | 90 | 0.7733 | |
| | | 0.7482 | 1.3 | 100 | 0.7694 | |
| | | 0.7366 | 1.43 | 110 | 0.7645 | |
| | | 0.7261 | 1.55 | 120 | 0.7605 | |
| | | 0.7171 | 1.68 | 130 | 0.7578 | |
| | | 0.7274 | 1.81 | 140 | 0.7552 | |
| | | 0.7313 | 1.94 | 150 | 0.7514 | |
| | | 0.7069 | 2.07 | 160 | 0.7516 | |
| | | 0.6858 | 2.2 | 170 | 0.7503 | |
| | | 0.6988 | 2.33 | 180 | 0.7498 | |
| | | 0.6903 | 2.46 | 190 | 0.7490 | |
| | | 0.6926 | 2.59 | 200 | 0.7490 | |
| | | 0.6952 | 2.72 | 210 | 0.7485 | |
| | | 0.6925 | 2.85 | 220 | 0.7484 | |
| | | 0.6938 | 2.98 | 230 | 0.7483 | |
| |
|
| |
|
| | ### Framework versions |
| |
|
| | - Transformers 4.36.0.dev0 |
| | - Pytorch 2.1.2+cu121 |
| | - Datasets 2.14.6 |
| | - Tokenizers 0.14.1 |
| |
|
