Crunchy-onion / README.md
leaderboard-pr-bot's picture
Adding Evaluation Results
f5edf0b verified
|
raw
history blame
3.64 kB
metadata
license: agpl-3.0
datasets:
  - lemonilia/LimaRP
  - grimulkan/theory-of-mind
  - Epiculous/Gnosis
model-index:
  - name: Crunchy-onion
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 67.15
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Epiculous/Crunchy-onion
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 86.19
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Epiculous/Crunchy-onion
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 70.02
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Epiculous/Crunchy-onion
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 63.88
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Epiculous/Crunchy-onion
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 73.24
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Epiculous/Crunchy-onion
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 52.01
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Epiculous/Crunchy-onion
          name: Open LLM Leaderboard

Crunchy-onion

This model is created by training Mixtral base model on LimaRP (ShareGPT format provided by SAO), theory of mind, and gnosis(provided by jeiku).

The 4-bit qlora was then merged into Mixtral Instruct resulting in what you see here.

Works best with Alpaca Instruct

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 68.75
AI2 Reasoning Challenge (25-Shot) 67.15
HellaSwag (10-Shot) 86.19
MMLU (5-Shot) 70.02
TruthfulQA (0-shot) 63.88
Winogrande (5-shot) 73.24
GSM8k (5-shot) 52.01