README.md

metadata

language:
  - en
license: cc-by-nc-4.0
datasets:
  - HuggingFaceH4/ultrafeedback_binarized
model-index:
  - name: loyal-piano-m7-cdpo
    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=chargoddard/loyal-piano-m7-cdpo
          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: 85.39
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/loyal-piano-m7-cdpo
          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: 64.52
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/loyal-piano-m7-cdpo
          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: 61.53
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/loyal-piano-m7-cdpo
          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: 79.4
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/loyal-piano-m7-cdpo
          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: 56.48
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/loyal-piano-m7-cdpo
          name: Open LLM Leaderboard

Trained for one epoch on ultrafeedback_binarized using cDPO. Evaluation pending.

Some initial benchmark results:

Task	Version	Metric	Value		Stderr
hellaswag	0	acc	0.6621	±	0.0047
		acc_norm	0.8525	±	0.0035
arc_challenge	0	acc	0.6348	±	0.0141
		acc_norm	0.6698	±	0.0137
winogrande	0	acc	0.7861	±	0.0115
gsm8k	0	acc	0.5694	±	0.0136

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	69.08
AI2 Reasoning Challenge (25-Shot)	67.15
HellaSwag (10-Shot)	85.39
MMLU (5-Shot)	64.52
TruthfulQA (0-shot)	61.53
Winogrande (5-shot)	79.40
GSM8k (5-shot)	56.48