README.md

metadata

license: apache-2.0
tags:
  - dpo
base_model:
  - mistralai/Mistral-7B-v0.1
dataset:
  - jondurbin/truthy-dpo-v0.1
model-index:
  - name: mistral-7b-jondurbin-truthy-dpo
    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: 60.75
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CorticalStack/mistral-7b-jondurbin-truthy-dpo
          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: 83.89
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CorticalStack/mistral-7b-jondurbin-truthy-dpo
          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: 63.65
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CorticalStack/mistral-7b-jondurbin-truthy-dpo
          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: 48.45
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CorticalStack/mistral-7b-jondurbin-truthy-dpo
          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: 77.98
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CorticalStack/mistral-7b-jondurbin-truthy-dpo
          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: 36.69
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=CorticalStack/mistral-7b-jondurbin-truthy-dpo
          name: Open LLM Leaderboard

mistral-7b-jondurbin-truthy-dpo

mistral-7b-jondurbin-truthy-dpo is a DPO fine-tuned version of mistralai/Mistral-7B-v0.1 using the jondurbin/truthy-dpo-v0.1 dataset.

LoRA

• r: 16
• LoRA alpha: 16
• LoRA dropout: 0.05

Training arguments

• Batch size: 4
• Gradient accumulation steps: 4
• Optimizer: paged_adamw_32bit
• Max steps: 100
• Learning rate: 5e-05
• Learning rate scheduler type: cosine
• Beta: 0.1
• Max prompt length: 1024
• Max length: 1536

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	61.90
AI2 Reasoning Challenge (25-Shot)	60.75
HellaSwag (10-Shot)	83.89
MMLU (5-Shot)	63.65
TruthfulQA (0-shot)	48.45
Winogrande (5-shot)	77.98
GSM8k (5-shot)	36.69