final_model-finetuned-20250514-162048

Fine-tuned Mistral Model for Clinical Tasks

Model Details

• Base Model: mistralai/Mistral-7B-Instruct-v0.2
• Fine-tuned on: MIMIC clinical data with instruction tuning
• Upload date: 2025-05-14
• Framework: PyTorch + Unsloth

Model Description

This model has been fine-tuned on MIMIC medical data with instruction tuning to improve clinical NLP capabilities such as ICD code extraction and medical information processing.

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "Pritish92/final_model-finetuned-20250514-162048" # This will be the HF repo ID like 'username/model-name'
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

# For simple text generation
inputs = tokenizer("Extract ICD-10 codes from this clinical note: ", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=200)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Training Information

• Training type: Instruction tuning with LoRA
• Hardware: Multiple GPUs: 2
• Original model: mistralai/Mistral-7B-Instruct-v0.2