Downloads · 30 days
5
5% of all-time downloads
nsumonte/higia-llama3.1-8b-medical-epoch2
higia-llama3.1-8b-medical-epoch2 is a machine learning model from nsumonte. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as llama3.1.
Modelo final especializado en razonamiento médico en español.
Downloads · 30 days
5
5% of all-time downloads
All-time downloads
95
Public
Parameters
266K
16.1 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors16.1 GB · 100%
From the Hugging Face model README
Modelo final especializado en razonamiento médico en español.
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("nsumonte/higia-llama3.1-8b-medical-epoch2", torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("nsumonte/higia-llama3.1-8b-medical-epoch2")
messages = [{"role": "user", "content": "¿Cuáles son los síntomas de diabetes tipo 2?"}]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
outputs = model.generate(inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0]))
| Epoch | Acc | Loss |
|---|---|---|
| 0 | 0.77 | 0.75 |
| 1 | 0.81 | 0.615 |
| 2 | 0.853 | 0.448 |
⚠️ NO reemplaza consejo médico profesional
Nicolás Sumonte - [email protected]