shawgpt-ft
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: 0.8247
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: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
3.8587 | 0.94 | 4 | 3.3972 |
3.1119 | 1.88 | 8 | 2.7188 |
2.4626 | 2.82 | 12 | 2.1773 |
1.5464 | 4.0 | 17 | 1.6636 |
1.5121 | 4.94 | 21 | 1.3225 |
1.1923 | 5.88 | 25 | 1.0879 |
0.9689 | 6.82 | 29 | 0.9322 |
0.7061 | 8.0 | 34 | 0.8552 |
0.8246 | 8.94 | 38 | 0.8292 |
0.7081 | 9.41 | 40 | 0.8247 |
Framework versions
- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 2
Model tree for Farisya/shawgpt-ft
Base model
mistralai/Mistral-7B-Instruct-v0.2
Quantized
TheBloke/Mistral-7B-Instruct-v0.2-GPTQ
Adapter
this model