metadata

library_name: transformers
tags:
  - mergekit
  - merge
base_model:
  - passing2961/Thanos-3B
  - prithivMLmods/Llama-Sentient-3.2-3B-Instruct
  - bunnycore/Llama-3.2-3B-RP-DeepThink
  - bunnycore/Llama-3.2-3B-All-Mix
  - huihui-ai/Llama-3.2-3B-Instruct-abliterated
  - prithivMLmods/Codepy-Deepthink-3B
  - HuggingFaceTB/finemath-ablation-infiwebmath
model-index:
  - name: Smol-Llama-3.2-3B
    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: 66.79
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Smol-Llama-3.2-3B
          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: 23.04
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Smol-Llama-3.2-3B
          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: 13.52
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Smol-Llama-3.2-3B
          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: 3.58
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Smol-Llama-3.2-3B
          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: 3.15
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Smol-Llama-3.2-3B
          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: 24.76
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=bunnycore/Smol-Llama-3.2-3B
          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 linear merge method using huihui-ai/Llama-3.2-3B-Instruct-abliterated as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:



merge_method: linear
dtype: bfloat16
normalize: true
base_model: huihui-ai/Llama-3.2-3B-Instruct-abliterated
models:
  - model: bunnycore/Llama-3.2-3B-All-Mix
    parameters:
      weight: 10
      density: 1
  - model: prithivMLmods/Codepy-Deepthink-3B
    parameters:
      weight: 7
      density: 0.8
  - model: huihui-ai/Llama-3.2-3B-Instruct-abliterated
    parameters:
      weight: 10
      density: 1
  - model: HuggingFaceTB/finemath-ablation-infiwebmath
    parameters:
      weight: 7
      density: 0.8
  - model: prithivMLmods/Llama-Sentient-3.2-3B-Instruct
    parameters:
      weight: 7
      density: 0.8
  - model: passing2961/Thanos-3B
    parameters:
      weight: 7
      density: 0.8
  - model: bunnycore/Llama-3.2-3B-RP-DeepThink
    parameters:
      weight: 7
      density: 0.8

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	22.47
IFEval (0-Shot)	66.79
BBH (3-Shot)	23.04
MATH Lvl 5 (4-Shot)	13.52
GPQA (0-shot)	3.58
MuSR (0-shot)	3.15
MMLU-PRO (5-shot)	24.76