README.md

metadata

library_name: transformers
tags:
  - mergekit
  - merge
base_model:
  - ZeroXClem/Qwen2.5-7B-HomerAnvita-NerdMix
  - ZeroXClem/Qwen2.5-7B-HomerCreative-Mix
model-index:
  - name: HomerCreativeAnvita-Mix-Qw7B
    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: 78.08
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/HomerCreativeAnvita-Mix-Qw7B
          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: 36.98
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/HomerCreativeAnvita-Mix-Qw7B
          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: 31.04
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/HomerCreativeAnvita-Mix-Qw7B
          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: 8.61
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/HomerCreativeAnvita-Mix-Qw7B
          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: 14.73
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/HomerCreativeAnvita-Mix-Qw7B
          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: 38.28
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/HomerCreativeAnvita-Mix-Qw7B
          name: Open LLM Leaderboard
license: apache-2.0

Merged Model

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

This model is currently ranked #3 on the Open LLM Leaderboard among models up to 8B parameters and #5 among models up to 13B parameters!

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

• ZeroXClem/Qwen2.5-7B-HomerAnvita-NerdMix
• ZeroXClem/Qwen2.5-7B-HomerCreative-Mix

Configuration

The following YAML configuration was used to produce this model:

base_model: ZeroXClem/Qwen2.5-7B-HomerCreative-Mix
dtype: bfloat16
merge_method: slerp
parameters:
  t:
  - filter: self_attn
    value: [0.0, 0.5, 0.3, 0.7, 1.0]
  - filter: mlp
    value: [1.0, 0.5, 0.7, 0.3, 0.0]
  - value: 0.5
slices:
- sources:
  - layer_range: [0, 28]
    model: ZeroXClem/Qwen2.5-7B-HomerCreative-Mix
  - layer_range: [0, 28]
    model: ZeroXClem/Qwen2.5-7B-HomerAnvita-NerdMix

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	34.62
IFEval (0-Shot)	78.08
BBH (3-Shot)	36.98
MATH Lvl 5 (4-Shot)	31.04
GPQA (0-shot)	8.61
MuSR (0-shot)	14.73
MMLU-PRO (5-shot)	38.28