Edit model card

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

train data

K RAG 6,123

mlflow

  • experiment : finetuning
  • run : SFT-PT_final-K

prompt template

### User: ๋‹น์‹ ์€ ์ธ๊ณต์ง€๋Šฅ ๋น„์„œ์ž…๋‹ˆ๋‹ค. ์‚ฌ์šฉ์ž๊ฐ€ ์—ฌ๋Ÿฌ๋ถ„์—๊ฒŒ ๊ณผ์ œ๋ฅผ ์ค๋‹ˆ๋‹ค. ๋‹น์‹ ์˜ ๋ชฉํ‘œ๋Š” ๊ฐ€๋Šฅํ•œ ํ•œ ์ถฉ์‹คํ•˜๊ฒŒ ์ž‘์—…์„ ์™„๋ฃŒํ•˜๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค. ์ž‘์—…์„ ์ˆ˜ํ–‰ํ•˜๋Š” ๋™์•ˆ ๋‹จ๊ณ„๋ณ„๋กœ ์ƒ๊ฐํ•˜๊ณ  ๋‹จ๊ณ„๋ฅผ ์ •๋‹นํ™”ํ•˜์„ธ์š”. User์˜ ์งˆ๋ฌธ์ด ์ฃผ์–ด์ง€๋ฉด ๊ณ ํ’ˆ์งˆ์˜ ๋‹ต๋ณ€์„ ๋งŒ๋“ค์–ด์ฃผ์„ธ์š”.
์›๋ฌธ: {CONTEXT}
์งˆ๋ฌธ: ์›๋ฌธ์„ ์ฐธ๊ณ ํ•˜์—ฌ ๋‹ต๋ณ€ํ•˜์„ธ์š”. {QUESTION}
 ### Assistant: {ANSWER}

train setting

  • Lora r, alpha : 32, 32
  • Dtype : bf16
  • Epoch : 5
  • Learning rate : 1e-5
  • Global batch : 1

inference setting

  • BOS id : 1
  • EOS id : 2
  • Top-p : 0.95
  • Temperature : 0.01

License

The content of this project, created by AGILESODA, is licensed under the Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Downloads last month
4
Safetensors
Model size
10.8B params
Tensor type
BF16
ยท
Inference API
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Collection including TwinDoc/agilesoda-model-x-sft-ver-k