README.md

metadata

license: apache-2.0
tags:
  - mergekit
  - moe
base_model:
  - nbeerbower/bruphin-epsilon
  - FelixChao/WestSeverus-7B-DPO-v2
  - jondurbin/airoboros-m-7b-3.1.2
model-index:
  - name: SuperBruphin-3x7B
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 71.16
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nbeerbower/SuperBruphin-3x7B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 87.74
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nbeerbower/SuperBruphin-3x7B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 64.58
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nbeerbower/SuperBruphin-3x7B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 66.85
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nbeerbower/SuperBruphin-3x7B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 81.53
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nbeerbower/SuperBruphin-3x7B
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 70.66
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nbeerbower/SuperBruphin-3x7B
          name: Open LLM Leaderboard

SuperBruphin-3x7B

This is an experimental MoE model created using mergekit. (mixtral branch)

Models Merged

The following models were included in the merge:

• nbeerbower/bruphin-epsilon
• FelixChao/WestSeverus-7B-DPO-v2
• jondurbin/airoboros-m-7b-3.1.2

Configuration

The following YAML configuration was used to produce this model:

base_model: nbeerbower/bruphin-epsilon
gate_mode: hidden
dtype: bfloat16
experts:
  - source_model: nbeerbower/bruphin-epsilon
    positive_prompts:
        - "Tell a story."
  - source_model: FelixChao/WestSeverus-7B-DPO-v2
    positive_prompts:
        - "Solve this problem."
  - source_model: jondurbin/airoboros-m-7b-3.1.2
    positive_prompts:
        - "Write a letter."

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	73.75
AI2 Reasoning Challenge (25-Shot)	71.16
HellaSwag (10-Shot)	87.74
MMLU (5-Shot)	64.58
TruthfulQA (0-shot)	66.85
Winogrande (5-shot)	81.53
GSM8k (5-shot)	70.66