README.md

metadata

license: apache-2.0
library_name: transformers
tags:
  - mergekit
  - merge
base_model:
  - TheDrummer/Big-Tiger-Gemma-27B-v1
  - Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
  - hotmailuser/Gemma2atlas-27B
  - hotmailuser/Gemma2SimPO-27B
model-index:
  - name: GemmaStock1-27B
    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: 75.09
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/GemmaStock1-27B
          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: 50.99
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/GemmaStock1-27B
          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: 25.91
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/GemmaStock1-27B
          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: 15.21
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/GemmaStock1-27B
          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: 15.99
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/GemmaStock1-27B
          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: 41.44
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=allknowingroger/GemmaStock1-27B
          name: Open LLM Leaderboard

merge

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the Model Stock merge method using Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B as a base.

Models Merged

The following models were included in the merge:

• TheDrummer/Big-Tiger-Gemma-27B-v1
• hotmailuser/Gemma2atlas-27B
• hotmailuser/Gemma2SimPO-27B

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: TheDrummer/Big-Tiger-Gemma-27B-v1
  - model: hotmailuser/Gemma2atlas-27B
  - model: hotmailuser/Gemma2SimPO-27B
  - model: Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
merge_method: model_stock
base_model: Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B
normalize: false
int8_mask: true
dtype: bfloat16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	37.44
IFEval (0-Shot)	75.09
BBH (3-Shot)	50.99
MATH Lvl 5 (4-Shot)	25.91
GPQA (0-shot)	15.21
MuSR (0-shot)	15.99
MMLU-PRO (5-shot)	41.44