Mistral-7b-sft-lora-ha-v1.0_cleaned
This model is a fine-tuned version of tokyotech-llm/Swallow-MS-7b-v0.1 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 1.2119
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: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.4836 | 1.0 | 189 | 1.2482 |
| 1.2209 | 2.0 | 378 | 1.2224 |
| 1.1941 | 3.0 | 567 | 1.2139 |
| 1.1819 | 4.0 | 756 | 1.2119 |
| 1.1787 | 5.0 | 945 | 1.2119 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 0
Unable to determine this model’s pipeline type. Check the
docs
.