ermwhatesigma420/sigmaAI:0.5b

ermwhatesigma420/ sigmaAI:0.5b

1 Download Updated 2 months ago

Self trained models. Based on instruct models.

0.5b

ollama run ermwhatesigma420/sigmaAI:0.5b

curl http://localhost:11434/api/chat \
  -d '{
    "model": "ermwhatesigma420/sigmaAI:0.5b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='ermwhatesigma420/sigmaAI:0.5b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'ermwhatesigma420/sigmaAI:0.5b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 2 months ago

2 months ago

a7c6d86a41f8 · 994MB ·

model

archqwen2

parameters494M

quantizationF16

994MB

system

You are Sigma AI, a based and helpful assistant. You speak with confidence and logic.

85B

params

{ "stop": [ "<|im_start|>", "<|im_end|>" ], "temperature": 0.4 }

77B

template

{{- range .Messages }}<|im_start|>{{ .Role }} {{ .Content }}<|im_end|> {{ end }}<|im_start|>assistan

102B

Readme

This is my first self trained ai model. It is based on the qwen 0.5b instruct. Trained on an json dataset. The dataset is uncensored. But i might not be the best model. It was a fun project making it tho. The 0.5 has a 20k json data set. Around 18mb, So a small dataset with the qwen small model.
The new sigmaAi