基于llama3 8B微调vicgalle/alpaca-gpt4数据集

ollama run fzkun/unsloth_llama3

curl http://localhost:11434/api/chat \
  -d '{
    "model": "fzkun/unsloth_llama3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='fzkun/unsloth_llama3',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'fzkun/unsloth_llama3',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 year ago

1 year ago

23f19d4ff55b · 8.5GB ·

model

archllama

parameters8.03B

quantizationQ8_0

8.5GB

template

Below are some instructions that describe some tasks. Write responses that appropriately complete ea

214B

params

{ "min_p": 0.1, "stop": [ "<|eot_id|>", "<|end_header_id|>", "<|star

187B

Readme

基于llama3(8B)进行微调

数据集(vicgalle/alpaca-gpt4数据集)

https://huggingface.co/datasets/vicgalle/alpaca-gpt4

colab地址：

https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Llama3_(8B)-Ollama.ipynb