metadata

language:
  - ko
  - en
  - jp
  - cn
license: apache-2.0
library_name: transformers
base_model: google/gemma-2-27b-it
datasets:
  - Saxo/ko_cn_translation_tech_social_science_linkbricks_single_dataset
  - Saxo/ko_jp_translation_tech_social_science_linkbricks_single_dataset
  - >-
    Saxo/en_ko_translation_tech_science_linkbricks_single_dataset_with_prompt_text_huggingface
  - >-
    Saxo/en_ko_translation_social_science_linkbricks_single_dataset_with_prompt_text_huggingface
  - >-
    Saxo/ko_aspect_sentiment_sns_mall_sentiment_linkbricks_single_dataset_with_prompt_text_huggingface
  - Saxo/ko_summarization_linkbricks_single_dataset_with_prompt_text_huggingface
  - >-
    Saxo/OpenOrca_cleaned_kor_linkbricks_single_dataset_with_prompt_text_huggingface
  - >-
    Saxo/ko_government_qa_total_linkbricks_single_dataset_with_prompt_text_huggingface_sampled
  - Saxo/ko-news-corpus-1
  - Saxo/ko-news-corpus-2
  - Saxo/ko-news-corpus-3
  - Saxo/ko-news-corpus-4
  - Saxo/ko-news-corpus-5
  - Saxo/ko-news-corpus-6
  - Saxo/ko-news-corpus-7
  - Saxo/ko-news-corpus-8
  - Saxo/ko-news-corpus-9
  - maywell/ko_Ultrafeedback_binarized
  - youjunhyeok/ko-orca-pair-and-ultrafeedback-dpo
  - lilacai/glaive-function-calling-v2-sharegpt
  - kuotient/gsm8k-ko
pipeline_tag: text-generation
model-index:
  - name: Linkbricks-Horizon-AI-Korean-Superb-27B
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 77.68
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 50.61
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 26.96
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 14.65
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 19.53
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 40.52
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
          name: Open LLM Leaderboard

Model Card for Model ID

AI 와 빅데이터 분석 전문 기업인 Linkbricks의 데이터사이언티스트인 지윤성(Saxo) 박사가
gemma-2-27b-it 베이스모델을 사용해서 H100-80G 8개를 통해 약 38%정도의 파라미터를 한국어 CPT(Continued-Pretraining)->SFT->DPO 한 한글 언어 모델
9천만건의 한글 뉴스 코퍼스를 기준으로 다양한 테스크별 한국어-중국어-영어-일본어 교차 학습 데이터와 수학 및 논리판단 데이터를 통하여 한중일영 언어 교차 증강 처리와 복잡한 논리 문제 역시 대응 가능하도록 훈련한 모델이다.
-토크나이저는 단어 확장 없이 베이스 모델 그대로 사용
-고객 리뷰나 소셜 포스팅 고차원 분석 및 코딩과 작문, 수학, 논리판단 등이 강화된 모델
-128k-Context Window
-Deepspeed Stage=3, rslora 및 BAdam Layer Mode 사용
"transformers_version": "4.46.1"

Finetuned by Mr. Yunsung Ji (Saxo), a data scientist at Linkbricks, a company specializing in AI and big data analytics
about 38% of total parameters Korean CPT(Continued-Pretraining)->SFT->DPO training model based on gemma-2-27b-it through 8 H100-80Gs as a Korean language model
It is a model that has been trained to handle Korean-Chinese-English-Japanese cross-training data and 90M korean news corpus and logic judgment data for various tasks to enable cross-fertilization processing and complex Korean logic & math problems.
-Tokenizer uses the base model without word expansion
-Models enhanced with high-dimensional analysis of customer reviews and social posts, as well as coding, writing, math and decision making
-128k-Context Window
-Deepspeed Stage=3, use rslora and BAdam Layer Mode

www.linkbricks.com, www.linkbricks.vc

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	38.32
IFEval (0-Shot)	77.68
BBH (3-Shot)	50.61
MATH Lvl 5 (4-Shot)	26.96
GPQA (0-shot)	14.65
MuSR (0-shot)	19.53
MMLU-PRO (5-shot)	40.52