midstream_llama_sft_lora_output

Medical LoRA adapter fine-tuned on clinical QA dataset with chain-of-thought reasoning.

Model Details

• Base model: meta-llama/Llama-3.1-8B-Instruct
• Adapter: LoRA (r=16, alpha=32)
• Training data: Medical case presentations with questions and answers
• Special: Chain-of-thought reasoning with <think> tags

Usage

from peft import PeftModel
from transformers import AutoTokenizer, AutoModelForCausalLM

base_model = "meta-llama/Llama-3.1-8B-Instruct"
adapter_id = "ofir408/midstream_llama_sft_lora_output"

tokenizer = AutoTokenizer.from_pretrained(base_model)
model = AutoModelForCausalLM.from_pretrained(base_model, device_map="auto")
model = PeftModel.from_pretrained(model, adapter_id)