buffettgpt-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: 0.6124
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 |
---|---|---|---|
2.031 | 0.992 | 31 | 0.8610 |
0.8683 | 1.984 | 62 | 0.6911 |
0.7677 | 2.976 | 93 | 0.5970 |
0.7038 | 4.0 | 125 | 0.5716 |
0.6997 | 4.992 | 156 | 0.5594 |
0.6701 | 5.984 | 187 | 0.5642 |
0.6402 | 6.976 | 218 | 0.5723 |
0.5945 | 8.0 | 250 | 0.5855 |
0.5881 | 8.992 | 281 | 0.6030 |
0.5596 | 9.92 | 310 | 0.6124 |
Framework versions
- PEFT 0.11.1
- Transformers 4.40.2
- Pytorch 2.1.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 0
Model tree for rawsem/buffettgpt-ft
Base model
mistralai/Mistral-7B-Instruct-v0.2
Quantized
TheBloke/Mistral-7B-Instruct-v0.2-GPTQ