ENERGY-DRINK-LOVE_-_komt_DPOv3-gguf – AI Model by RichardErkhov | AlphaNeural AI

You can deploy this model and start earning money today!

RichardErkhov/
ENERGY-DRINK-LOVE_-_komt_DPOv3-gguf

like

0

endpoints_compatible

Views

No views yet

Quantization made by Richard Erkhov.

Request more models

komt_DPOv3 - GGUF

Model creator: https://huggingface.co/ENERGY-DRINK-LOVE/
Original model: https://huggingface.co/ENERGY-DRINK-LOVE/komt_DPOv3/

Name	Quant method	Size
komt_DPOv3.Q2_K.gguf	Q2_K	3.8GB
komt_DPOv3.Q3_K_S.gguf	Q3_K_S	4.42GB
komt_DPOv3.Q3_K.gguf	Q3_K	4.92GB
komt_DPOv3.Q3_K_M.gguf	Q3_K_M	4.92GB
komt_DPOv3.Q3_K_L.gguf	Q3_K_L	5.34GB
komt_DPOv3.IQ4_XS.gguf	IQ4_XS	5.51GB
komt_DPOv3.Q4_0.gguf	Q4_0	5.74GB
komt_DPOv3.IQ4_NL.gguf	IQ4_NL	5.8GB
komt_DPOv3.Q4_K_S.gguf	Q4_K_S	5.78GB
komt_DPOv3.Q4_K.gguf	Q4_K	6.1GB
komt_DPOv3.Q4_K_M.gguf	Q4_K_M	6.1GB
komt_DPOv3.Q4_1.gguf	Q4_1	6.36GB
komt_DPOv3.Q5_0.gguf	Q5_0	6.98GB
komt_DPOv3.Q5_K_S.gguf	Q5_K_S	6.98GB
komt_DPOv3.Q5_K.gguf	Q5_K	7.17GB
komt_DPOv3.Q5_K_M.gguf	Q5_K_M	7.17GB
komt_DPOv3.Q5_1.gguf	Q5_1	7.6GB
komt_DPOv3.Q6_K.gguf	Q6_K	8.3GB
komt_DPOv3.Q8_0.gguf	Q8_0	10.75GB

Original model description:

license: cc-by-4.0 base_model: davidkim205/komt-solar-10.7b-sft-v5 tags:

trl
dpo
generated_from_trainer model-index:
name: nhn_dpo_v3_komt-solar-10.7b-sft-v5_DPO results: []

ENERGY-DRINK-LOVE/komt_DPOv3

Our Team

Youjin Chung
Jingyeom Kim

Model

Base Model

davidkim205/komt-solar-10.7b-sft-v5

Hardware and Software

Hardware: A100 * 8 for training our model
Deepspeed library & Huggingface TRL Trainer

Dataset

DPO_dataset
- 자체 제작 dpo dataset(AI-hub dataset 활용)
- OpenOrca DPO 등 영어 데이터셋 번역(ENERGY-DRINK-LOVE/translate_share_gpt_dedup_llama_SFT_1024, 자체모델 활용)

Training Method

DPO

Benchmark

Ko LM Eval Harness

Ko-LLM-Leaderboard

(240316기준 4등)
image/png

Average	Ko-ARC	Ko-HellaSwag	Ko-MMLU	Ko-TruthfulQA	Ko-CommonGen V2
61.20	57.51	70.33	53.34	68.49	56.32