Qwen3-0.6B-v0.1
DeepSpeed-Chat์ผ๋ก ํ์ธํ๋๋ ์ธ์ด ๋ชจ๋ธ
Model Details
์ด ๋ชจ๋ธ์ DeepSpeed-Chat์ ์ฌ์ฉํ์ฌ ํ์ธํ๋๋ ๋ชจ๋ธ์ ๋๋ค.
- Base Model: ๊ธฐ๋ณธ ๋ชจ๋ธ ์ ๋ณด๋ฅผ ์ฌ๊ธฐ์ ์ถ๊ฐํ์ธ์
- Fine-tuning Method: DeepSpeed-Chat
- Training Data: ํ์ต ๋ฐ์ดํฐ ์ ๋ณด๋ฅผ ์ฌ๊ธฐ์ ์ถ๊ฐํ์ธ์
Usage
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("mncai/Qwen3-0.6B-v0.1")
model = AutoModelForCausalLM.from_pretrained("mncai/Qwen3-0.6B-v0.1")
# ํ
์คํธ ์์ฑ
input_text = "Your prompt here"
inputs = tokenizer(input_text, return_tensors="pt")
outputs = model.generate(**inputs, max_length=100)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
Training Details
- Training Framework: DeepSpeed
- Training Script: DeepSpeed-Chat Step 1 Supervised Fine-tuning
- Upload Date: N/A
Limitations and Biases
์ด ๋ชจ๋ธ์ ํ๊ณ์ ๊ณผ ํธํฅ์ฑ์ ๋ํ ์ ๋ณด๋ฅผ ์ฌ๊ธฐ์ ์ถ๊ฐํ์ธ์.
Citation
DeepSpeed-Chat์ ์ฌ์ฉํ๋ค๋ฉด ๋ค์์ ์ธ์ฉํด์ฃผ์ธ์:
@misc{deepspeed-chat,
title={DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scales},
author={Yuxiao Zhuang et al.},
year={2023},
url={https://github.com/microsoft/DeepSpeed}
}
- Downloads last month
- 18