Gefeliciteerd!
Werken met Llama 3
Imtihan Ahmed
Machine Learning Engineer
Even opfrissen
from llama_cpp import Llama
llm = Llama(model_path = "path/to/model.gguf")
Even opfrissen
$$
temperature,top_k,top_p-parameters
Even opfrissen
message_list = [{"role": "system", "content": system_message},
{"role": "user", "content": user_message}]
Even opfrissen
- Precieze prompts
stop-woorden- Zero-shot/few-shot prompting
Even opfrissen
$$
response_format = {"type": "json_object"}
Even opfrissen
Conversation-klasse.create_completion()-methode
