lastmass/llama3.2-chinese

lastmass/ llama3.2-chinese:latest

504 Downloads Updated 1 year ago

llama3.2 countinue pretrain with wiki-zh and sft with chinese

ollama run lastmass/llama3.2-chinese

curl http://localhost:11434/api/chat \
  -d '{
    "model": "lastmass/llama3.2-chinese",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='lastmass/llama3.2-chinese',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'lastmass/llama3.2-chinese',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 year ago

1 year ago

fa0e70eac0b6 · 2.2GB ·

model

archllama

parameters3.61B

quantizationQ4_K_M

2.2GB

params

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

275B

template

以下是描述一项任务的指令。请输出合适的内容回应指令。 ### Input: 大象和�

893B

Readme

Continued pretraining with wikimedia/wikipedia 20231101.zh Instruction Finetuning with FreedomIntelligence/alpaca-gpt4-chinese

run huggingface model ollama run hf.co/lastmass/llama3.2-chinese:Q4_K_M