README.md

metadata

base_model:
  - johnsutor/Llama-3-8B-Instruct_dare_ties-density-0.9
  - FuseAI/FuseChat-Llama-3.1-8B-SFT
  - refuelai/Llama-3-Refueled
  - DreadPoor/Derivative-8B-Model_Stock
  - Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base
library_name: transformers
tags:
  - mergekit
  - merge
model-index:
  - name: Spring_Dusk-8B-SCE
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: wis-k/instruction-following-eval
          split: train
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 65.15
            name: averaged accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=DreadPoor%2FSpring_Dusk-8B-SCE
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: SaylorTwift/bbh
          split: test
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 37.76
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=DreadPoor%2FSpring_Dusk-8B-SCE
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: lighteval/MATH-Hard
          split: test
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 7.4
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=DreadPoor%2FSpring_Dusk-8B-SCE
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          split: train
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 5.03
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=DreadPoor%2FSpring_Dusk-8B-SCE
          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: 17.33
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=DreadPoor%2FSpring_Dusk-8B-SCE
          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: 27.06
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=DreadPoor%2FSpring_Dusk-8B-SCE
          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 SCE merge method using FuseAI/FuseChat-Llama-3.1-8B-SFT as a base.

Models Merged

The following models were included in the merge:

• johnsutor/Llama-3-8B-Instruct_dare_ties-density-0.9
• refuelai/Llama-3-Refueled
• DreadPoor/Derivative-8B-Model_Stock
• Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: refuelai/Llama-3-Refueled
  - model: johnsutor/Llama-3-8B-Instruct_dare_ties-density-0.9
  - model: Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base
  - model: DreadPoor/Derivative-8B-Model_Stock
merge_method: sce
base_model: FuseAI/FuseChat-Llama-3.1-8B-SFT
parameters:
  select_topk: 0.3
dtype: bfloat16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here! Summarized results can be found here!

Metric	Value (%)
Average	26.62
IFEval (0-Shot)	65.15
BBH (3-Shot)	37.76
MATH Lvl 5 (4-Shot)	7.40
GPQA (0-shot)	5.03
MuSR (0-shot)	17.33
MMLU-PRO (5-shot)	27.06