README.md

metadata

library_name: transformers
tags:
  - mergekit
  - merge
base_model:
  - shuttleai/shuttle-2.5-mini
  - mistralai/Mistral-Nemo-Base-2407
  - Undi95/LocalC-12B-e2.0
model-index:
  - name: StableProse
    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: 19.72
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nlpguy/StableProse
          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: 30.18
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nlpguy/StableProse
          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: 4.68
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nlpguy/StableProse
          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: 7.05
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nlpguy/StableProse
          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: 8.87
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nlpguy/StableProse
          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: 27.43
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nlpguy/StableProse
          name: Open LLM Leaderboard

EDIT: This Merge inherits bad IFEval Results from the Base Model, keep this in mind when using this model.

merged

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 mistralai/Mistral-Nemo-Base-2407 as a base.

Models Merged

The following models were included in the merge:

• shuttleai/shuttle-2.5-mini
• Undi95/LocalC-12B-e2.0

Configuration

The following YAML configuration was used to produce this model:

base_model: mistralai/Mistral-Nemo-Base-2407
dtype: bfloat16
merge_method: model_stock
slices:
- sources:
  - layer_range: [0, 40]
    model: mistralai/Mistral-Nemo-Base-2407
  - layer_range: [0, 40]
    model: Undi95/LocalC-12B-e2.0
  - layer_range: [0, 40]
    model: shuttleai/shuttle-2.5-mini

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	16.32
IFEval (0-Shot)	19.72
BBH (3-Shot)	30.18
MATH Lvl 5 (4-Shot)	4.68
GPQA (0-shot)	7.05
MuSR (0-shot)	8.87
MMLU-PRO (5-shot)	27.43