BioMistral_DS_fine_tuned

This model is a fine-tuned version of BioMistral/BioMistral-7B on the daphne604/Mic_mortality_reason dataset. It achieves the following results on the evaluation set:

• Loss: 0.5240

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss
1.3883	0.9964	137	1.2905
1.0024	2.0	275	0.8735
0.4672	2.9964	412	0.6598
0.3044	4.0	550	0.5674
0.2501	4.9964	687	0.5263
0.5557	5.9782	822	0.5240

Framework versions

• PEFT 0.13.2
• Transformers 4.46.3
• Pytorch 2.5.0
• Datasets 3.0.1
• Tokenizers 0.20.1

Cite TRL as:

@misc{BioMistral_fine_tuned,
    title = {daphne604/{B}io{M}istral\_{D}{S}\_fine\_tuned · {H}ugging {F}ace --- huggingface.co},
    author       = {Daphne},
    year = {2024},
    publisher = {Hugging Face},
    journal = {Hugging Face repository},
    howpublished =  {\url{https://huggingface.co/daphne604/BioMistral_DS_fine_tuned}}
}

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}