README.md

---
license: cc-by-nc-4.0
datasets:
- pankajmathur/orca_mini_v1_dataset
- openai/summarize_from_feedback
- PygmalionAI/PIPPA
- chargoddard/rpguild
- lemonilia/LimaRP
- PKU-Alignment/PKU-SafeRLHF
- Intel/orca_dpo_pairs
- allenai/ultrafeedback_binarized_cleaned
model-index:
- name: servile-harpsichord-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: 67.32
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/servile-harpsichord-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: 85.18
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/servile-harpsichord-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: 64.54
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/servile-harpsichord-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: 60.61
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/servile-harpsichord-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: 79.16
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/servile-harpsichord-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: 57.09
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/servile-harpsichord-cdpo
      name: Open LLM Leaderboard
---

Trained on a different random sampling of the same datasets used by [loyal-piano-m7](https://huggingface.co/chargoddard/loyal-piano-m7), then with cDPO on a blend of RLHF datasets.

Several intermediate checkpoints (of cDPO training) are on branches.

Uses the Alpaca prompt format.
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_chargoddard__servile-harpsichord-cdpo)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |68.98|
|AI2 Reasoning Challenge (25-Shot)|67.32|
|HellaSwag (10-Shot)              |85.18|
|MMLU (5-Shot)                    |64.54|
|TruthfulQA (0-shot)              |60.61|
|Winogrande (5-shot)              |79.16|
|GSM8k (5-shot)                   |57.09|