Tilldela chattroller

Arbeta med Llama 3

Imtihan Ahmed

Machine Learning Engineer

Definiera roller

  • Justera modellens ton
  • Exempel: kundtjänst-chatbot

Konversation med en chatbot - 1

Arbeta med Llama 3

Definiera roller

  • Justera modellens ton
  • Exempel: kundtjänst-chatbot

Konversation med en chatbot - 2

Arbeta med Llama 3

Definiera roller

  • Justera modellens ton
  • Exempel: kundtjänst-chatbot

Konversation med en chatbot - 3

Arbeta med Llama 3

Använda roller i chattgenerering

  • Chattroller styr Llamas svar

Screenshot 2025-02-26 at 15.32.28.png

  • Anger personlighet och stil

 

Screenshot 2025-02-26 at 15.31.50.png

  • Representerar den som ställer frågan
Arbeta med Llama 3

Använda roller i chattgenerering

  • Skicka en strukturerad konversation

  • Funktionen create_chat_completion()

from llama_cpp import Llama

llm = Llama(model_path="path/to/model.gguf")


message_list = [...] # This list includes roles


response = llm.create_chat_completion(

        messages = message_list

)
Arbeta med Llama 3

Rollen system

  • Meddelandet System: instruktioner om hur modellen ska bete sig
system_message = "You are a business consultant who gives data-driven answers."



message_list = [{

                 "role": "system",

                 "content": system_message

                 }]
Arbeta med Llama 3

Rollen user

  • Meddelandet User: prompten som skickas till modellen
system_message = "You are a business consultant who gives data-driven answers."

user_message = "What are the key factors in a successful marketing strategy?"


message_list = [{"role": "system", "content": system_message},

                {
                 "role": "user",
                 "content": user_message
                 }

                ]
Arbeta med Llama 3

Generera svaret

from llama_cpp import Llama
llm = Llama(model_path="path/to/model.gguf")

system_message = "You are a business consultant who gives data-driven answers."
user_message = "What are the key factors in a successful marketing strategy?"

message_list = [{"role": "system", "content": system_message},
               {"role": "user", "content": user_message}]


response = llm.create_chat_completion(messages = message_list)
print(response)

{'id': ..., 'object': ..., 'created': ..., 'model': ..., 'choices': [...], ...}
Arbeta med Llama 3

Rollen assistant

response["choices"][0]

Svarsutdata

Arbeta med Llama 3

Rollen assistant

response["choices"][0]

Svarsutdata med meddelandet markerat

Arbeta med Llama 3

Rollen assistant

response["choices"][0]

Svarsutdata med rollen markerad

Arbeta med Llama 3

Rollen assistant

response["choices"][0]

Svarsutdata med innehållet markerat

result['choices'][0]['message']['content']

'A successful marketing strategy relies on clear objectives, established 
through specific, measurable goals. Understanding the target audience ...'
Arbeta med Llama 3

Nu kör vi en övning!

Arbeta med Llama 3

Preparing Video For Download...