nbeerbower's picture
Adding Evaluation Results (#1)
4e0c886 verified
|
raw
history blame
3.92 kB
metadata
license: gemma
library_name: transformers
base_model:
  - UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3
datasets:
  - jondurbin/gutenberg-dpo-v0.1
  - nbeerbower/gutenberg2-dpo
model-index:
  - name: Gemma2-Gutenberg-Doppel-9B
    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: 71.71
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nbeerbower/Gemma2-Gutenberg-Doppel-9B
          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: 41.08
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nbeerbower/Gemma2-Gutenberg-Doppel-9B
          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: 3.47
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nbeerbower/Gemma2-Gutenberg-Doppel-9B
          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: 10.63
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nbeerbower/Gemma2-Gutenberg-Doppel-9B
          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: 17.3
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nbeerbower/Gemma2-Gutenberg-Doppel-9B
          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: 34.75
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=nbeerbower/Gemma2-Gutenberg-Doppel-9B
          name: Open LLM Leaderboard

image/png

Gemma2-Gutenberg-Doppel-9B

UCLA-AGI/Gemma-2-9B-It-SPPO-Iter3 finetuned on jondurbin/gutenberg-dpo-v0.1 and nbeerbower/gutenberg2-dpo.

Method

ORPO finetuned using 2x A40 for 3 epochs.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 29.82
IFEval (0-Shot) 71.71
BBH (3-Shot) 41.08
MATH Lvl 5 (4-Shot) 3.47
GPQA (0-shot) 10.63
MuSR (0-shot) 17.30
MMLU-PRO (5-shot) 34.75