| --- |
| library_name: peft |
| license: llama3.1 |
| base_model: unsloth/llama-3.1-8b-instruct-unsloth-bnb-4bit |
| tags: |
| - trl |
| - sft |
| - unsloth |
| - generated_from_trainer |
| model-index: |
| - name: outputs |
| 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. --> |
|
|
| # outputs |
|
|
| This model is a fine-tuned version of [unsloth/llama-3.1-8b-instruct-unsloth-bnb-4bit](https://huggingface.co/unsloth/llama-3.1-8b-instruct-unsloth-bnb-4bit) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 1.3917 |
|
|
| ## 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: 2e-05 |
| - train_batch_size: 2 |
| - eval_batch_size: 8 |
| - seed: 3407 |
| - gradient_accumulation_steps: 16 |
| - total_train_batch_size: 32 |
| - optimizer: Use adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
| - lr_scheduler_type: linear |
| - lr_scheduler_warmup_ratio: 0.05 |
| - num_epochs: 3 |
|
|
| ### Training results |
|
|
| | Training Loss | Epoch | Step | Validation Loss | |
| |:-------------:|:------:|:----:|:---------------:| |
| | 1.8249 | 0.1014 | 10 | 1.7483 | |
| | 1.7082 | 0.2028 | 20 | 1.6279 | |
| | 1.6314 | 0.3042 | 30 | 1.5369 | |
| | 1.5673 | 0.4056 | 40 | 1.4923 | |
| | 1.5182 | 0.5070 | 50 | 1.4698 | |
| | 1.4977 | 0.6084 | 60 | 1.4576 | |
| | 1.4762 | 0.7098 | 70 | 1.4489 | |
| | 1.4641 | 0.8112 | 80 | 1.4424 | |
| | 1.4763 | 0.9125 | 90 | 1.4365 | |
| | 1.5972 | 1.0203 | 100 | 1.4317 | |
| | 1.4753 | 1.1217 | 110 | 1.4267 | |
| | 1.4513 | 1.2231 | 120 | 1.4224 | |
| | 1.4775 | 1.3245 | 130 | 1.4188 | |
| | 1.4559 | 1.4259 | 140 | 1.4156 | |
| | 1.4214 | 1.5272 | 150 | 1.4126 | |
| | 1.4534 | 1.6286 | 160 | 1.4098 | |
| | 1.4176 | 1.7300 | 170 | 1.4073 | |
| | 1.4204 | 1.8314 | 180 | 1.4046 | |
| | 1.4398 | 1.9328 | 190 | 1.4027 | |
| | 1.5789 | 2.0406 | 200 | 1.4010 | |
| | 1.4298 | 2.1420 | 210 | 1.3990 | |
| | 1.4299 | 2.2433 | 220 | 1.3974 | |
| | 1.4136 | 2.3447 | 230 | 1.3959 | |
| | 1.4535 | 2.4461 | 240 | 1.3949 | |
| | 1.4244 | 2.5475 | 250 | 1.3938 | |
| | 1.4079 | 2.6489 | 260 | 1.3929 | |
| | 1.4013 | 2.7503 | 270 | 1.3922 | |
| | 1.4204 | 2.8517 | 280 | 1.3918 | |
| | 1.4224 | 2.9531 | 290 | 1.3917 | |
|
|
|
|
| ### Framework versions |
|
|
| - PEFT 0.15.2 |
| - Transformers 4.51.3 |
| - Pytorch 2.6.0+cu124 |
| - Datasets 3.5.0 |
| - Tokenizers 0.21.1 |
