CHF0101/medquad-lora-r32-best-v2

MedQuAD LoRA r=32

Configuración

• Base: mistralai/Mistral-7B-Instruct-v0.3
• LoRA r: 32
• Módulos: q_proj, k_proj, v_proj
• 4-bit NF4
• Early Stopping: patience=3

Entrenamiento

Training logs (manual, Epoch estimado, 150 max steps):

Step	Epoch	Training Loss	Validation Loss
50	0.023	0.882800	0.845063
100	0.046	0.805700	0.784524
150	0.070	0.754400	0.767385

Evaluación

• BertScore F1 (sample test 50): 0.8452

Uso

from peft import PeftModel
from transformers import AutoModelForCausalLM
base = AutoModelForCausalLM.from_pretrained('mistralai/Mistral-7B-Instruct-v0.3', load_in_4bit=True)
model = PeftModel.from_pretrained(base, 'CHF0101/medquad-lora-r32-best-v2')