taejoonlee/v4

taejoonlee/ v4:latest

5 Downloads Updated 1 year ago

Llama 3 8B + 7K Dataset, templete modified version

ollama run taejoonlee/v4

curl http://localhost:11434/api/chat \
  -d '{
    "model": "taejoonlee/v4",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='taejoonlee/v4',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'taejoonlee/v4',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 1 year ago

1 year ago

f2126df5c52d · 8.5GB ·

model

archllama

parameters8.03B

quantizationQ8_0

8.5GB

params

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

187B

template

Below are some instructions that describe some tasks. Start your response with a JSON object, where

417B

Readme

chat_template = “”“Below are some instructions that describe some tasks. Start your response with a JSON object, followed by a plain text answer.

Instruction:

{INPUT}

Response:

{{“is_recipe_request”: {is_recipe_request}, “recipe_name”: “{recipe_name}”}} {OUTPUT}

”“”

from unsloth import apply_chat_template dataset = apply_chat_template( dataset, tokenizer = tokenizer, chat_template = chat_template, default_system_message=“Start your response with a JSON object including ‘is_recipe_request’ and ‘recipe_name’. Follow with a plain text answer detailing the response.” )