metadata
base_model: mistralai/Mistral-7B-Instruct-v0.3
datasets:
- GaetanMichelet/chat-60_ft_task-1
- GaetanMichelet/chat-120_ft_task-1
library_name: peft
license: apache-2.0
tags:
- alignment-handbook
- trl
- sft
- generated_from_trainer
model-index:
- name: Mistral-7B_task-1_120-samples_config-1
results: []
Mistral-7B_task-1_120-samples_config-1
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the GaetanMichelet/chat-60_ft_task-1 and the GaetanMichelet/chat-120_ft_task-1 datasets. It achieves the following results on the evaluation set:
- Loss: 1.1514
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.8479 | 1.0 | 11 | 1.7767 |
| 1.3159 | 2.0 | 22 | 1.3060 |
| 1.0602 | 3.0 | 33 | 1.1514 |
| 0.7949 | 4.0 | 44 | 1.1724 |
| 0.5137 | 5.0 | 55 | 1.3468 |
| 0.2333 | 6.0 | 66 | 1.6914 |
| 0.1177 | 7.0 | 77 | 1.8391 |
| 0.0738 | 8.0 | 88 | 1.8831 |
| 0.0676 | 9.0 | 99 | 1.9784 |
| 0.0588 | 10.0 | 110 | 2.1311 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1