| | --- |
| | license: other |
| | base_model: yahma/llama-7b-hf |
| | tags: |
| | - generated_from_trainer |
| | model-index: |
| | - name: V0305P8 |
| | 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. --> |
| |
|
| | # V0305P8 |
| |
|
| | 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: 0.0762 |
| |
|
| | ## 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 | |
| | |:-------------:|:-----:|:----:|:---------------:| |
| | | 1.877 | 0.09 | 10 | 0.8565 | |
| | | 0.3259 | 0.17 | 20 | 0.1568 | |
| | | 0.1567 | 0.26 | 30 | 0.1546 | |
| | | 0.152 | 0.34 | 40 | 0.1487 | |
| | | 0.1493 | 0.43 | 50 | 0.1432 | |
| | | 0.149 | 0.51 | 60 | 0.1259 | |
| | | 0.1321 | 0.6 | 70 | 0.1074 | |
| | | 0.1234 | 0.68 | 80 | 0.1031 | |
| | | 0.1094 | 0.77 | 90 | 0.0987 | |
| | | 0.11 | 0.85 | 100 | 0.0989 | |
| | | 0.1084 | 0.94 | 110 | 0.0971 | |
| | | 0.1055 | 1.02 | 120 | 0.0967 | |
| | | 0.0937 | 1.11 | 130 | 0.0920 | |
| | | 0.0917 | 1.19 | 140 | 0.0874 | |
| | | 0.0893 | 1.28 | 150 | 0.0819 | |
| | | 0.09 | 1.37 | 160 | 0.0808 | |
| | | 0.0882 | 1.45 | 170 | 0.0827 | |
| | | 0.0832 | 1.54 | 180 | 0.0830 | |
| | | 0.0824 | 1.62 | 190 | 0.0798 | |
| | | 0.0871 | 1.71 | 200 | 0.0800 | |
| | | 0.0861 | 1.79 | 210 | 0.0792 | |
| | | 0.0844 | 1.88 | 220 | 0.0761 | |
| | | 0.0802 | 1.96 | 230 | 0.0780 | |
| | | 0.0691 | 2.05 | 240 | 0.0805 | |
| | | 0.0602 | 2.13 | 250 | 0.0816 | |
| | | 0.056 | 2.22 | 260 | 0.0802 | |
| | | 0.0544 | 2.3 | 270 | 0.0834 | |
| | | 0.0632 | 2.39 | 280 | 0.0769 | |
| | | 0.0629 | 2.47 | 290 | 0.0735 | |
| | | 0.0615 | 2.56 | 300 | 0.0739 | |
| | | 0.0625 | 2.65 | 310 | 0.0756 | |
| | | 0.056 | 2.73 | 320 | 0.0764 | |
| | | 0.0586 | 2.82 | 330 | 0.0761 | |
| | | 0.0576 | 2.9 | 340 | 0.0763 | |
| | | 0.0582 | 2.99 | 350 | 0.0762 | |
| |
|
| |
|
| | ### Framework versions |
| |
|
| | - Transformers 4.36.0.dev0 |
| | - Pytorch 2.1.2+cu121 |
| | - Datasets 2.14.6 |
| | - Tokenizers 0.14.1 |
| |
|
