README.md

metadata

license: llama3.3
library_name: transformers
tags:
  - mergekit
  - merge
base_model:
  - TheDrummer/Anubis-70B-v1
  - TheDrummer/Nautilus-70B-v0.1
  - codelion/Llama-3.3-70B-o1
model-index:
  - name: Set-70b
    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: 76.43
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Set-70b
          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: 56.88
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Set-70b
          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: 36.33
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Set-70b
          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: 26.17
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Set-70b
          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: 18.96
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Set-70b
          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: 49.36
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Set-70b
          name: Open LLM Leaderboard

Merge

This is a merge of pre-trained language models created using mergekit.

Merge Details

RP with some o1 inspiration.

Merge Method

This model was merged using the Model Stock merge method using codelion/Llama-3.3-70B-o1 as a base.

Models Merged

The following models were included in the merge:

• TheDrummer/Anubis-70B-v1
• TheDrummer/Nautilus-70B-v0.1

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: codelion/Llama-3.3-70B-o1
  - model: TheDrummer/Anubis-70B-v1
  - model: TheDrummer/Nautilus-70B-v0.1
base_model: codelion/Llama-3.3-70B-o1
merge_method: model_stock
parameters:
  normalize: true
dtype: bfloat16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	44.02
IFEval (0-Shot)	76.43
BBH (3-Shot)	56.88
MATH Lvl 5 (4-Shot)	36.33
GPQA (0-shot)	26.17
MuSR (0-shot)	18.96
MMLU-PRO (5-shot)	49.36