README.md

metadata

language:
  - en
license: mit
library_name: transformers
tags:
  - mergekit
  - merge
  - phi-4
base_model:
  - bunnycore/Phi-4-RR-Shoup
  - bunnycore/Phi-4-Model-Stock-v4
pipeline_tag: text-generation
model-index:
  - name: Luminis-phi-4
    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: 69
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/Luminis-phi-4
          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: 55.8
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/Luminis-phi-4
          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: 43.66
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/Luminis-phi-4
          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: 13.53
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/Luminis-phi-4
          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: 16.68
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/Luminis-phi-4
          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: 49.15
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=suayptalha/Luminis-phi-4
          name: Open LLM Leaderboard

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 15B parameters, #4 among models up to 32B parameters and #53 among all the models!

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

• bunnycore/Phi-4-RR-Shoup
• bunnycore/Phi-4-Model-Stock-v4

Configuration

The following YAML configuration was used to produce this model:

base_model: bunnycore/Phi-4-Model-Stock-v4
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, 40]
    model: bunnycore/Phi-4-Model-Stock-v4
  - layer_range: [0, 40]
    model: bunnycore/Phi-4-RR-Shoup

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	41.30
IFEval (0-Shot)	69.00
BBH (3-Shot)	55.80
MATH Lvl 5 (4-Shot)	43.66
GPQA (0-shot)	13.53
MuSR (0-shot)	16.68
MMLU-PRO (5-shot)	49.15