---
license: apache-2.0
base_model: mistralai/Mistral-7B-Instruct-v0.2
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: UTI_L3_1000steps_1e6rate_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. -->

# UTI_L3_1000steps_1e6rate_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: 3.2252

## 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-06
- 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 |
|:-------------:|:-------:|:----:|:---------------:|
| 2.2806        | 0.3333  | 25   | 2.1563          |
| 1.8955        | 0.6667  | 50   | 1.8431          |
| 1.7404        | 1.0     | 75   | 1.7251          |
| 1.5351        | 1.3333  | 100  | 1.7245          |
| 1.5176        | 1.6667  | 125  | 1.7039          |
| 1.4594        | 2.0     | 150  | 1.6754          |
| 0.9453        | 2.3333  | 175  | 1.8231          |
| 1.0165        | 2.6667  | 200  | 1.8012          |
| 0.9754        | 3.0     | 225  | 1.7995          |
| 0.4991        | 3.3333  | 250  | 2.1098          |
| 0.4971        | 3.6667  | 275  | 2.1236          |
| 0.5339        | 4.0     | 300  | 2.1092          |
| 0.2532        | 4.3333  | 325  | 2.2857          |
| 0.2919        | 4.6667  | 350  | 2.3444          |
| 0.3192        | 5.0     | 375  | 2.3734          |
| 0.1858        | 5.3333  | 400  | 2.5514          |
| 0.1947        | 5.6667  | 425  | 2.5828          |
| 0.1984        | 6.0     | 450  | 2.5324          |
| 0.1429        | 6.3333  | 475  | 2.7141          |
| 0.1573        | 6.6667  | 500  | 2.6237          |
| 0.1502        | 7.0     | 525  | 2.6715          |
| 0.1168        | 7.3333  | 550  | 2.8434          |
| 0.1306        | 7.6667  | 575  | 2.7996          |
| 0.1182        | 8.0     | 600  | 2.8128          |
| 0.1009        | 8.3333  | 625  | 2.9270          |
| 0.1053        | 8.6667  | 650  | 2.9832          |
| 0.0983        | 9.0     | 675  | 2.9935          |
| 0.0887        | 9.3333  | 700  | 3.0662          |
| 0.0894        | 9.6667  | 725  | 3.0845          |
| 0.0914        | 10.0    | 750  | 3.0977          |
| 0.0829        | 10.3333 | 775  | 3.1662          |
| 0.0775        | 10.6667 | 800  | 3.1832          |
| 0.0841        | 11.0    | 825  | 3.1821          |
| 0.0753        | 11.3333 | 850  | 3.2082          |
| 0.078         | 11.6667 | 875  | 3.2170          |
| 0.0745        | 12.0    | 900  | 3.2223          |
| 0.0788        | 12.3333 | 925  | 3.2260          |
| 0.0743        | 12.6667 | 950  | 3.2258          |
| 0.0718        | 13.0    | 975  | 3.2253          |
| 0.0744        | 13.3333 | 1000 | 3.2252          |


### Framework versions

- Transformers 4.41.1
- Pytorch 2.0.0+cu117
- Datasets 2.19.1
- Tokenizers 0.19.1