README.md

metadata

base_model:
  - T145/ZEUS-8B-V17
  - Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2
library_name: transformers
tags:
  - mergekit
  - merge
model-index:
  - name: ZEUS-8B-V17-abliterated
    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: 75.76
            name: averaged accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=T145%2FZEUS-8B-V17-abliterated
          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: 31.52
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=T145%2FZEUS-8B-V17-abliterated
          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: 2.27
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=T145%2FZEUS-8B-V17-abliterated
          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: 7.16
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=T145%2FZEUS-8B-V17-abliterated
          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: 13.13
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=T145%2FZEUS-8B-V17-abliterated
          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: 29.13
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=T145%2FZEUS-8B-V17-abliterated
          name: Open LLM Leaderboard

ZEUS-8B-V17-abliterated

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

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

• T145/ZEUS-8B-V17
• Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2

Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
    - model: T145/ZEUS-8B-V17
      layer_range: [0, 18]
  - sources:
    # Reasoning: The script used to abliterate V2 & V13 identified layer 19 as the target with the most refusal.
    # Substituting the whole layer with that from the uncensored model should effectively abliterate V17 and future similar merges.
    - model: Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2
      layer_range: [19, 19]
  - sources:
    - model: T145/ZEUS-8B-V17
      layer_range: [20, 32]
merge_method: passthrough
tokenizer_source: T145/ZEUS-8B-V17
dtype: bfloat16

Open LLM Leaderboard Evaluation Results

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

Metric	Value (%)
Average	26.50
IFEval (0-Shot)	75.76
BBH (3-Shot)	31.52
MATH Lvl 5 (4-Shot)	2.27
GPQA (0-shot)	7.16
MuSR (0-shot)	13.13
MMLU-PRO (5-shot)	29.13