final_model-3
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.0656
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
The following bitsandbytes
quantization config was used during training:
- quant_method: bitsandbytes
- load_in_8bit: False
- load_in_4bit: True
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: nf4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float16
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- training_steps: 20
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.4774 | 0.0 | 1 | 2.3519 |
2.3907 | 0.01 | 2 | 2.2619 |
2.2831 | 0.01 | 3 | 2.1934 |
2.3523 | 0.02 | 4 | 2.1535 |
2.2008 | 0.02 | 5 | 2.1415 |
2.398 | 0.02 | 6 | 2.1336 |
2.0323 | 0.03 | 7 | 2.1223 |
1.9787 | 0.03 | 8 | 2.1102 |
2.2163 | 0.04 | 9 | 2.1011 |
2.4075 | 0.04 | 10 | 2.0942 |
2.0822 | 0.04 | 11 | 2.0878 |
2.3128 | 0.05 | 12 | 2.0823 |
1.9674 | 0.05 | 13 | 2.0775 |
2.0991 | 0.06 | 14 | 2.0739 |
2.1918 | 0.06 | 15 | 2.0707 |
2.0037 | 0.06 | 16 | 2.0684 |
2.0398 | 0.07 | 17 | 2.0669 |
2.1113 | 0.07 | 18 | 2.0661 |
1.9206 | 0.08 | 19 | 2.0657 |
1.6649 | 0.08 | 20 | 2.0656 |
Framework versions
- PEFT 0.4.0
- Transformers 4.37.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.15.2
- Downloads last month
- 0
Model tree for hussamsal/final_model-3
Base model
mistralai/Mistral-7B-Instruct-v0.2