metadata
license: apache-2.0
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
base_model: mistralai/Mistral-7B-Instruct-v0.1
model-index:
- name: Mistral-7B-Medical-Finetune_QA
results: []
Mistral-7B-Medical-Finetune_QA
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6753
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: 0.00025
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8304 | 0.15 | 300 | 0.7822 |
| 0.7735 | 0.29 | 600 | 0.7528 |
| 0.7522 | 0.44 | 900 | 0.7275 |
| 0.7234 | 0.59 | 1200 | 0.7043 |
| 0.7048 | 0.74 | 1500 | 0.6850 |
| 0.6894 | 0.88 | 1800 | 0.6753 |
Framework versions
- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2