|
--- |
|
license: apache-2.0 |
|
base_model: mistralai/Mistral-7B-Instruct-v0.1 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: mistral-try-finetune |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# mistral-try-finetune |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3805 |
|
|
|
## 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: 2.5e-05 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 8 |
|
- seed: 18 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: constant |
|
- lr_scheduler_warmup_steps: 5 |
|
- training_steps: 1000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.5606 | 0.57 | 50 | 0.8581 | |
|
| 0.5656 | 1.14 | 100 | 0.5153 | |
|
| 0.3651 | 1.71 | 150 | 0.4257 | |
|
| 0.2995 | 2.29 | 200 | 0.3750 | |
|
| 0.2008 | 2.86 | 250 | 0.3405 | |
|
| 0.1693 | 3.43 | 300 | 0.3282 | |
|
| 0.144 | 4.0 | 350 | 0.3156 | |
|
| 0.1112 | 4.57 | 400 | 0.3209 | |
|
| 0.0949 | 5.14 | 450 | 0.3346 | |
|
| 0.0801 | 5.71 | 500 | 0.3212 | |
|
| 0.0717 | 6.29 | 550 | 0.3288 | |
|
| 0.0579 | 6.86 | 600 | 0.3255 | |
|
| 0.0486 | 7.43 | 650 | 0.3359 | |
|
| 0.0495 | 8.0 | 700 | 0.3273 | |
|
| 0.0374 | 8.57 | 750 | 0.3617 | |
|
| 0.0377 | 9.14 | 800 | 0.3725 | |
|
| 0.0324 | 9.71 | 850 | 0.3697 | |
|
| 0.0338 | 10.29 | 900 | 0.3946 | |
|
| 0.0305 | 10.86 | 950 | 0.3605 | |
|
| 0.0289 | 11.43 | 1000 | 0.3805 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
