---
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: mistralai/Mistral-7B-Instruct-v0.2
datasets:
- generator
metrics:
- bleu
- rouge
model-index:
- name: Mistral-7B-Instruct-v0.2-advisegpt-v0.6
  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-Instruct-v0.2-advisegpt-v0.6

This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0767
- Bleu: {'bleu': 0.9584832765902116, 'precisions': [0.9778312591422885, 0.9625878953932084, 0.9518774970032065, 0.9430684559898991], 'brevity_penalty': 0.9997177244264667, 'length_ratio': 0.9997177642587203, 'translation_length': 1289338, 'reference_length': 1289702}
- Rouge: {'rouge1': 0.9761023152523122, 'rouge2': 0.9590922549283836, 'rougeL': 0.9747297976860183, 'rougeLsum': 0.9758442544146716}
- Exact Match: {'exact_match': 0.0}

## 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: 3
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 10
- total_train_batch_size: 30
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 3
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Bleu                                                                                                                                                                                                                                                                | Rouge                                                                                                                       | Exact Match          |
|:-------------:|:------:|:----:|:---------------:|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------------------------:|:--------------------:|
| 0.067         | 0.9998 | 809  | 0.0945          | {'bleu': 0.9492918853166353, 'precisions': [0.9733554685833311, 0.9543042005762523, 0.9412361771045687, 0.9307382413966919], 'brevity_penalty': 0.9994904502180469, 'length_ratio': 0.9994905799944483, 'translation_length': 1289045, 'reference_length': 1289702} | {'rouge1': 0.9712558044405124, 'rouge2': 0.9500703853191179, 'rougeL': 0.9690578078497468, 'rougeLsum': 0.9708044674114953} | {'exact_match': 0.0} |
| 0.0527        | 1.9995 | 1618 | 0.0779          | {'bleu': 0.9568445996007577, 'precisions': [0.977026202258449, 0.961055539100332, 0.9498195483213825, 0.9405540074014527], 'brevity_penalty': 0.9998193217903225, 'length_ratio': 0.9998193381106644, 'translation_length': 1289469, 'reference_length': 1289702}   | {'rouge1': 0.9753094821779227, 'rouge2': 0.9574822736836266, 'rougeL': 0.9737984768450723, 'rougeLsum': 0.9750220632065946} | {'exact_match': 0.0} |
| 0.0471        | 2.9993 | 2427 | 0.0767          | {'bleu': 0.9584832765902116, 'precisions': [0.9778312591422885, 0.9625878953932084, 0.9518774970032065, 0.9430684559898991], 'brevity_penalty': 0.9997177244264667, 'length_ratio': 0.9997177642587203, 'translation_length': 1289338, 'reference_length': 1289702} | {'rouge1': 0.9761023152523122, 'rouge2': 0.9590922549283836, 'rougeL': 0.9747297976860183, 'rougeLsum': 0.9758442544146716} | {'exact_match': 0.0} |


### Framework versions

- PEFT 0.10.0
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1