¡Enhorabuena!
Trabajar con Llama 3
Imtihan Ahmed
Machine Learning Engineer
Recordemos
from llama_cpp import Llama
llm = Llama(model_path = "path/to/model.gguf")
Recordemos
$$
- Parámetros
temperature,top_k,top_p
Recordemos
message_list = [{"role": "system", "content": system_message},
{"role": "user", "content": user_message}]
Recordemos
- Prompts precisos
- Palabras de
stop - Zero-shot/Few-shot prompting
Recordemos
$$
response_format = {"type": "json_object"}
Recordemos
- Clase
Conversation - Método
.create_completion()
