metadata
language:
- ko
- en
library_name: transformers
license: cc-by-nc-sa-4.0
pipeline_tag: text-generation
tags:
- pytorch
Model Description
S-B κ³ κ°μ¬ νλ‘μ νΈ μ μμ±ν RAG λ°μ΄ν°μ μ νμ©νμ¬ Supervised Fine-Tuning(a.k.a SFT) νμ΅ν λͺ¨λΈμ λλ€. νμ΅ λ°μ΄ν°μ μ 보μμ μν΄ κ³΅κ°νμ§ μμ΅λλ€.
About the Model
Name: TwinDoc/RedWhale-tv-10.8B-sft-s
Finetuned from model: TwinDoc/RedWhale-tv-10.8B-v1.0
Train Datasets: private
Developed by: μ μμΌμλ€ (AGILESODA)
Model type: llama
Language(s) (NLP): νκ΅μ΄
License: cc-by-nc-sa-4.0
train setting
- Lora r, alpha : 4, 16
- Dtype : bf16
- Epoch : 7
- Learning rate : 1e-4
- Global batch : 4
- Context length : 4096
inference setting
- BOS id : 1
- EOS id : 2
- Top-p : 0.95
- Temperature : 0.01
prompt template
Human: ##μλ¬Έ##κ³Ό ##μ§λ¬Έ##μ΄ μ£Όμ΄μ§λ©΄, ##μλ¬Έ##μ μλ μ 보λ₯Ό λ°νμΌλ‘ κ³ νμ§μ ##λ΅λ³##μ λ§λ€μ΄μ£ΌμΈμ. ##μλ¬Έ##μμ ##μ§λ¬Έ##μ λν λͺ
νν λ΅μ μ°Ύμ μ μμ κ²½μ° "λ΅λ³μ μ°Ύμ μ μμ΅λλ€."λ‘ ##λ΅λ³##μ μμ±ν΄μΌνλ©° ##μλ¬Έ##μ μλ λ΄μ©μ ##λ΅λ³##μ ν¬ν¨νμ§ μμμΌ ν©λλ€.
##μλ¬Έ##
{CONTEXT}
##μ§λ¬Έ##
{QUESTION}
Assistant: {ANSWER}
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).
Citation
@misc{vo2024redwhaleadaptedkoreanllm,
title={RedWhale: An Adapted Korean LLM Through Efficient Continual Pretraining},
author={Anh-Dung Vo and Minseong Jung and Wonbeen Lee and Daewoo Choi},
year={2024},
eprint={2408.11294},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2408.11294},
}
Built with: