metadata
base_model: mistralai/Mistral-7B-Instruct-v0.3
datasets:
- GaetanMichelet/chat-60_ft_task-1_auto
- GaetanMichelet/chat-120_ft_task-1_auto
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-2_auto
results: []
Mistral-7B_task-1_120-samples_config-2_auto
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.3 on the GaetanMichelet/chat-60_ft_task-1_auto and the GaetanMichelet/chat-120_ft_task-1_auto datasets. It achieves the following results on the evaluation set:
- Loss: 0.8006
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: 16
- total_train_batch_size: 16
- 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 |
|---|---|---|---|
| 1.0732 | 0.9091 | 5 | 1.0410 |
| 0.9533 | 2.0 | 11 | 0.8652 |
| 0.77 | 2.9091 | 16 | 0.8181 |
| 0.6091 | 4.0 | 22 | 0.8006 |
| 0.4937 | 4.9091 | 27 | 0.8277 |
| 0.3093 | 6.0 | 33 | 0.9964 |
| 0.1731 | 6.9091 | 38 | 1.0815 |
| 0.1045 | 8.0 | 44 | 1.3044 |
| 0.0642 | 8.9091 | 49 | 1.3456 |
| 0.0552 | 10.0 | 55 | 1.4001 |
| 0.0359 | 10.9091 | 60 | 1.4895 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.0
- Pytorch 2.1.2+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1