shawgpt-ft-ep15-r16
This model is a fine-tuned version of TheBloke/Mistral-7B-Instruct-v0.2-GPTQ on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.3611
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: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.5907 | 0.9231 | 3 | 3.9581 |
| 4.0319 | 1.8462 | 6 | 3.4176 |
| 3.4291 | 2.7692 | 9 | 2.9301 |
| 2.1958 | 4.0 | 13 | 2.4634 |
| 2.5294 | 4.9231 | 16 | 2.1678 |
| 2.1123 | 5.8462 | 19 | 1.8719 |
| 1.8035 | 6.7692 | 22 | 1.7037 |
| 1.2441 | 8.0 | 26 | 1.5414 |
| 1.5017 | 8.9231 | 29 | 1.4422 |
| 1.3882 | 9.8462 | 32 | 1.4007 |
| 1.38 | 10.7692 | 35 | 1.3818 |
| 0.9914 | 12.0 | 39 | 1.3685 |
| 1.3368 | 12.9231 | 42 | 1.3632 |
| 1.1111 | 13.8462 | 45 | 1.3611 |
Framework versions
- PEFT 0.13.2
- Transformers 4.44.2
- Pytorch 2.5.0+cu121
- Datasets 3.0.2
- Tokenizers 0.19.1
- Downloads last month
- 4
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for GabrielG228/shawgpt-ft-ep15-r16
Base model
mistralai/Mistral-7B-Instruct-v0.2 Quantized
TheBloke/Mistral-7B-Instruct-v0.2-GPTQ