README.md

metadata

license: apache-2.0
tags:
  - merge
model-index:
  - name: ColorShadow-7B-v3
    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: 67.58
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/ColorShadow-7B-v3
          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: 85.04
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/ColorShadow-7B-v3
          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: 60.57
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/ColorShadow-7B-v3
          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: 62.88
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/ColorShadow-7B-v3
          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: 80.11
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/ColorShadow-7B-v3
          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: 47.54
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=nlpguy/ColorShadow-7B-v3
          name: Open LLM Leaderboard

ColorShadow-7B-v3

This is a Gradient-SLERP merge between ColorShadow-7B and Terminis-7B performed using mergekit.

Here is the config file used:

slices:
  - sources:
      - model: nlpguy/ColorShadow-7B
        layer_range: [0, 32]
      - model: Q-bert/Terminis-7B
        layer_range: [0, 32]
merge_method: slerp
base_model: nlpguy/ColorShadow-7B
parameters:
  t:
    - filter: self_attn
      value: [1, 0.5, 0.7, 0.3, 0]
    - filter: mlp
      value: [0, 0.5, 0.3, 0.7, 1]
    - value: 0.5 # fallback for rest of tensors
dtype: float16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	67.29
AI2 Reasoning Challenge (25-Shot)	67.58
HellaSwag (10-Shot)	85.04
MMLU (5-Shot)	60.57
TruthfulQA (0-shot)	62.88
Winogrande (5-shot)	80.11
GSM8k (5-shot)	47.54