Robo-Gutenberg_V1.0 / README.md
Triangle104's picture
Adding Evaluation Results (#1)
d451582 verified
metadata
license: apache-2.0
library_name: transformers
tags:
  - mergekit
  - merge
base_model:
  - huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2
  - rombodawg/Rombos-LLM-V2.6-Qwen-14b
  - nbeerbower/Qwen2.5-Gutenberg-Doppel-14B
model-index:
  - name: Robo-Gutenberg_V1.0
    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: 60.08
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Robo-Gutenberg_V1.0
          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: 50.29
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Robo-Gutenberg_V1.0
          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: 21.3
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Robo-Gutenberg_V1.0
          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: 18.12
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Robo-Gutenberg_V1.0
          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: 19.2
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Robo-Gutenberg_V1.0
          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: 48.79
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=Triangle104/Robo-Gutenberg_V1.0
          name: Open LLM Leaderboard

merge

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


Model details:

image/webp

He'll be back... or something.

Quants can be found here:

https://huggingface.co/mradermacher/Robo-Gutenberg_V1.0-GGUF

https://huggingface.co/mradermacher/Robo-Gutenberg_V1.0-i1-GGUF


Merge Method

This model was merged using the TIES merge method using rombodawg/Rombos-LLM-V2.6-Qwen-14b as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: rombodawg/Rombos-LLM-V2.6-Qwen-14b
    #no parameters necessary for base model
  - model: nbeerbower/Qwen2.5-Gutenberg-Doppel-14B
    parameters:
      density: 0.7
      weight: 0.7
  - model: huihui-ai/Qwen2.5-14B-Instruct-abliterated-v2
    parameters:
      density: 0.5
      weight: 0.5

merge_method: ties
base_model: rombodawg/Rombos-LLM-V2.6-Qwen-14b
parameters:
  normalize: false
  int8_mask: true
dtype: float16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 36.30
IFEval (0-Shot) 60.08
BBH (3-Shot) 50.29
MATH Lvl 5 (4-Shot) 21.30
GPQA (0-shot) 18.12
MuSR (0-shot) 19.20
MMLU-PRO (5-shot) 48.79