|
--- |
|
license: apache-2.0 |
|
base_model: mistralai/Mistral-7B-Instruct-v0.2 |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
model-index: |
|
- name: Transaminitis_M2_1000rate_1e5_SFT |
|
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. --> |
|
|
|
# Transaminitis_M2_1000rate_1e5_SFT |
|
|
|
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co./mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3335 |
|
|
|
## 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: 1e-05 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 4 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- training_steps: 1000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.481 | 0.2 | 25 | 0.6212 | |
|
| 0.4843 | 0.4 | 50 | 0.4602 | |
|
| 0.4738 | 0.6 | 75 | 0.7299 | |
|
| 0.5545 | 0.8 | 100 | 0.8047 | |
|
| 0.5804 | 1.0 | 125 | 0.2867 | |
|
| 0.3273 | 1.2 | 150 | 0.4552 | |
|
| 0.4405 | 1.4 | 175 | 0.2827 | |
|
| 0.3231 | 1.6 | 200 | 0.3794 | |
|
| 0.26 | 1.8 | 225 | 0.2519 | |
|
| 0.2472 | 2.0 | 250 | 0.2499 | |
|
| 0.2571 | 2.2 | 275 | 0.2425 | |
|
| 0.2316 | 2.4 | 300 | 0.2324 | |
|
| 0.2285 | 2.6 | 325 | 0.2367 | |
|
| 0.2325 | 2.8 | 350 | 0.2349 | |
|
| 0.2273 | 3.0 | 375 | 0.2316 | |
|
| 0.2202 | 3.2 | 400 | 0.2258 | |
|
| 0.2178 | 3.4 | 425 | 0.2277 | |
|
| 0.2228 | 3.6 | 450 | 0.2267 | |
|
| 0.2213 | 3.8 | 475 | 0.2246 | |
|
| 0.2173 | 4.0 | 500 | 0.2363 | |
|
| 0.2097 | 4.2 | 525 | 0.2388 | |
|
| 0.2123 | 4.4 | 550 | 0.2288 | |
|
| 0.2139 | 4.6 | 575 | 0.2252 | |
|
| 0.2162 | 4.8 | 600 | 0.2245 | |
|
| 0.2108 | 5.0 | 625 | 0.2249 | |
|
| 0.1981 | 5.2 | 650 | 0.2304 | |
|
| 0.2025 | 5.4 | 675 | 0.2325 | |
|
| 0.1982 | 5.6 | 700 | 0.2335 | |
|
| 0.1966 | 5.8 | 725 | 0.2322 | |
|
| 0.2021 | 6.0 | 750 | 0.2314 | |
|
| 0.1772 | 6.2 | 775 | 0.2624 | |
|
| 0.1733 | 6.4 | 800 | 0.2670 | |
|
| 0.169 | 6.6 | 825 | 0.2719 | |
|
| 0.1682 | 6.8 | 850 | 0.2800 | |
|
| 0.177 | 7.0 | 875 | 0.2782 | |
|
| 0.1408 | 7.2 | 900 | 0.3174 | |
|
| 0.1398 | 7.4 | 925 | 0.3290 | |
|
| 0.142 | 7.6 | 950 | 0.3331 | |
|
| 0.1336 | 7.8 | 975 | 0.3337 | |
|
| 0.1378 | 8.0 | 1000 | 0.3335 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.2 |
|
- Pytorch 2.0.0+cu117 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
