A newer version of this model is available: CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct-2412

image/webp

Model Description

Model Details

  • Name: Carrot Llama-3.2 Rabbit Ko
  • Version: 3B Instruct
  • Base Model: CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct
  • Languages: Korean, English
  • Model Type: Large Language Model (Instruction-tuned)

Training Process

๋ณธ ๋ชจ๋ธ์€ ๋‹ค์Œ๊ณผ ๊ฐ™์€ ์ฃผ์š” ํ›ˆ๋ จ ๋‹จ๊ณ„๋ฅผ ๊ฑฐ์ณค์Šต๋‹ˆ๋‹ค:

  1. SFT (Supervised Fine-Tuning)
    • ๊ณ ํ’ˆ์งˆ ํ•œ๊ตญ์–ด ๋ฐ ์˜์–ด ๋ฐ์ดํ„ฐ์…‹์„ ์‚ฌ์šฉํ•˜์—ฌ ๊ธฐ๋ณธ ๋ชจ๋ธ์„ ์„ธ๋ถ€ ์กฐ์ •

Limitations

  • 3B ํŒŒ๋ผ๋ฏธํ„ฐ ๊ทœ๋ชจ๋กœ ์ธํ•œ ๋ณต์žกํ•œ ์ž‘์—…์—์„œ์˜ ์ œํ•œ์  ์„ฑ๋Šฅ
  • ํŠน์ • ๋„๋ฉ”์ธ์— ๋Œ€ํ•œ ๊นŠ์ด ์žˆ๋Š” ์ „๋ฌธ์„ฑ ๋ถ€์กฑ
  • ํŽธํ–ฅ์„ฑ ๋ฐ ํ™˜๊ฐ ๊ฐ€๋Šฅ์„ฑ

Ethics Statement

๋ชจ๋ธ ๊ฐœ๋ฐœ ๊ณผ์ •์—์„œ ์œค๋ฆฌ์  ๊ณ ๋ ค์‚ฌํ•ญ์„ ์ตœ๋Œ€ํ•œ ๋ฐ˜์˜ํ•˜์˜€์œผ๋‚˜, ์‚ฌ์šฉ์ž๋Š” ํ•ญ์ƒ ๊ฒฐ๊ณผ๋ฅผ ๋น„ํŒ์ ์œผ๋กœ ๊ฒ€ํ† ํ•ด์•ผ ํ•ฉ๋‹ˆ๋‹ค.

How to Use

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct")
tokenizer = AutoTokenizer.from_pretrained("CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct")

Score

Tasks Version Filter n-shot Metric Value Stderr
gsm8k 3 flexible-extract 5 exact_match โ†‘ 0.6490 ยฑ 0.0131
strict-match 5 exact_match โ†‘ 0.0023 ยฑ 0.0013
gsm8k-ko 3 flexible-extract 5 exact_match โ†‘ 0.3275 ยฑ 0.0134
strict-match 5 exact_match โ†‘ 0.2737 ยฑ 0.0134
ifeval 4 none 5 inst_level_loose_acc โ†‘ 0.8058 ยฑ N/A
none 5 inst_level_strict_acc โ†‘ 0.7686 ยฑ N/A
none 5 prompt_level_loose_acc โ†‘ 0.7320 ยฑ 0.0191
none 5 prompt_level_strict_acc โ†‘ 0.6858 ยฑ 0.0200
Tasks Version Filter n-shot Metric Value Stderr
haerae 1 none acc โ†‘ 0.4180 ยฑ 0.0148
none acc_norm โ†‘ 0.4180 ยฑ 0.0148
- haerae_general_knowledge 1 none 5 acc โ†‘ 0.3125 ยฑ 0.0350
none 5 acc_norm โ†‘ 0.3125 ยฑ 0.0350
- haerae_history 1 none 5 acc โ†‘ 0.3404 ยฑ 0.0347
none 5 acc_norm โ†‘ 0.3404 ยฑ 0.0347
- haerae_loan_word 1 none 5 acc โ†‘ 0.4083 ยฑ 0.0379
none 5 acc_norm โ†‘ 0.4083 ยฑ 0.0379
- haerae_rare_word 1 none 5 acc โ†‘ 0.4815 ยฑ 0.0249
none 5 acc_norm โ†‘ 0.4815 ยฑ 0.0249
- haerae_standard_nomenclature 1 none 5 acc โ†‘ 0.4771 ยฑ 0.0405
none 5 acc_norm โ†‘ 0.4771 ยฑ 0.0405
Tasks Version Filter n-shot Metric Value Stderr
kobest_boolq 1 none 5 acc โ†‘ 0.7664 ยฑ 0.0113
none 5 f1 โ†‘ 0.7662 ยฑ N/A
kobest_copa 1 none 5 acc โ†‘ 0.5620 ยฑ 0.0157
none 5 f1 โ†‘ 0.5612 ยฑ N/A
kobest_hellaswag 1 none 5 acc โ†‘ 0.3840 ยฑ 0.0218
none 5 acc_norm โ†‘ 0.4900 ยฑ 0.0224
none 5 f1 โ†‘ 0.3807 ยฑ N/A
kobest_sentineg 1 none 5 acc โ†‘ 0.5869 ยฑ 0.0247
none 5 f1 โ†‘ 0.5545 ยฑ N/A
kobest_wic 1 none 5 acc โ†‘ 0.4952 ยฑ 0.0141
none 5 f1 โ†‘ 0.4000 ยฑ N/A
@article{Llama3.2RabbitKo3BInstruct,
  title={CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct Card},
  author={CarrotAI (L, GEUN)},
  year={2024},
  url = {https://huggingface.co/CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct}
}
Downloads last month
2,391
Safetensors
Model size
3.21B params
Tensor type
BF16
ยท
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.

Model tree for CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct

Finetuned
(246)
this model
Finetunes
1 model
Merges
4 models
Quantizations
4 models

Dataset used to train CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct

Collection including CarrotAI/Llama-3.2-Rabbit-Ko-3B-Instruct