| --- |
| license: other |
| library_name: peft |
| tags: |
| - generated_from_trainer |
| base_model: google/gemma-2b |
| model-index: |
| - name: Gemma2B-StaproCoder |
| 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. --> |
|
|
| # Gemma2B-StaproCoder |
|
|
| This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.2988 |
|
|
| ## 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: 0.0005 |
| - train_batch_size: 1 |
| - eval_batch_size: 1 |
| - seed: 42 |
| - gradient_accumulation_steps: 4 |
| - total_train_batch_size: 4 |
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| - lr_scheduler_type: cosine |
| - lr_scheduler_warmup_ratio: 0.1 |
| - training_steps: 2000 |
|
|
| ### Training results |
|
|
| | Training Loss | Epoch | Step | Validation Loss | |
| |:-------------:|:-----:|:----:|:---------------:| |
| | 0.733 | 0.05 | 100 | 0.5294 | |
| | 0.5149 | 0.1 | 200 | 0.4275 | |
| | 0.2925 | 0.15 | 300 | 0.3854 | |
| | 0.3588 | 0.2 | 400 | 0.3794 | |
| | 0.3145 | 0.25 | 500 | 0.3766 | |
| | 0.4036 | 0.3 | 600 | 0.3728 | |
| | 0.4822 | 0.35 | 700 | 0.3553 | |
| | 0.3456 | 0.4 | 800 | 0.3428 | |
| | 0.3978 | 0.45 | 900 | 0.3367 | |
| | 0.2692 | 0.5 | 1000 | 0.3365 | |
| | 0.4038 | 0.55 | 1100 | 0.3203 | |
| | 0.3345 | 0.6 | 1200 | 0.3210 | |
| | 0.2668 | 0.65 | 1300 | 0.3130 | |
| | 0.2617 | 0.7 | 1400 | 0.3103 | |
| | 0.2657 | 0.75 | 1500 | 0.3099 | |
| | 0.2633 | 0.8 | 1600 | 0.3041 | |
| | 0.4033 | 0.85 | 1700 | 0.3045 | |
| | 0.2208 | 0.9 | 1800 | 0.3017 | |
| | 0.2646 | 0.95 | 1900 | 0.2989 | |
| | 0.3054 | 1.0 | 2000 | 0.2988 | |
|
|
|
|
| ### Framework versions |
|
|
| - PEFT 0.9.0 |
| - Transformers 4.39.0.dev0 |
| - Pytorch 2.2.1+cu121 |
| - Datasets 2.17.1 |
| - Tokenizers 0.15.2 |
