|
--- |
|
base_model: mistralai/Mistral-7B-Instruct-v0.3 |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: mistral-7b-peptide |
|
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-7b-peptide |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.5836 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 3 |
|
- eval_batch_size: 3 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 48 |
|
- total_eval_batch_size: 12 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 30 |
|
- training_steps: 4000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 2.7541 | 0.025 | 100 | 2.6109 | |
|
| 2.2057 | 0.05 | 200 | 2.1101 | |
|
| 1.9566 | 0.075 | 300 | 1.8904 | |
|
| 1.8737 | 0.1 | 400 | 3.6582 | |
|
| 1.8384 | 0.125 | 500 | 1.6622 | |
|
| 1.6577 | 0.15 | 600 | 1.6209 | |
|
| 5.0415 | 0.175 | 700 | 5.0107 | |
|
| 4.8597 | 0.2 | 800 | 4.8365 | |
|
| 4.7887 | 0.225 | 900 | 4.7727 | |
|
| 4.7478 | 0.25 | 1000 | 4.7247 | |
|
| 4.8828 | 0.275 | 1100 | 4.8448 | |
|
| 4.5764 | 0.3 | 1200 | 4.4572 | |
|
| 4.2131 | 0.325 | 1300 | 4.1309 | |
|
| 3.8945 | 0.35 | 1400 | 3.7905 | |
|
| 3.42 | 0.375 | 1500 | 3.2011 | |
|
| 1.6361 | 0.4 | 1600 | 1.5822 | |
|
| 1.4804 | 0.425 | 1700 | 1.5127 | |
|
| 1.3574 | 0.45 | 1800 | 1.5037 | |
|
| 1.2675 | 0.475 | 1900 | 1.4394 | |
|
| 1.2611 | 0.5 | 2000 | 1.3705 | |
|
| 1.1509 | 0.525 | 2100 | 1.3520 | |
|
| 1.0144 | 0.55 | 2200 | 1.3529 | |
|
| 1.3122 | 0.575 | 2300 | 1.2730 | |
|
| 1.0257 | 0.6 | 2400 | 1.2805 | |
|
| 0.7651 | 0.625 | 2500 | 1.3131 | |
|
| 0.5841 | 0.65 | 2600 | 1.3736 | |
|
| 0.4848 | 0.675 | 2700 | 1.4138 | |
|
| 0.6076 | 0.7 | 2800 | 1.3322 | |
|
| 0.4255 | 0.725 | 2900 | 1.4169 | |
|
| 0.3276 | 0.75 | 3000 | 1.4631 | |
|
| 0.6833 | 0.775 | 3100 | 1.2651 | |
|
| 0.385 | 0.8 | 3200 | 1.3994 | |
|
| 0.1845 | 0.825 | 3300 | 1.4685 | |
|
| 0.1408 | 0.85 | 3400 | 1.5640 | |
|
| 0.1213 | 0.875 | 3500 | 1.5984 | |
|
| 0.1735 | 0.9 | 3600 | 1.5952 | |
|
| 0.1161 | 0.925 | 3700 | 1.6201 | |
|
| 0.1079 | 0.95 | 3800 | 1.6238 | |
|
| 0.3046 | 0.975 | 3900 | 1.6070 | |
|
| 0.1477 | 1.0 | 4000 | 1.5836 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.0.dev0 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.20.0 |
|
- Tokenizers 0.19.1 |
|
|
