---
license: mit
library_name: transformers
base_model: wandb/mistral-7b-zephyr-sft
datasets:
- argilla/dpo-mix-7k
model-index:
- name: mistral-7b-zephyr-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: 63.05
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=wandb/mistral-7b-zephyr-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: 85.54
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=wandb/mistral-7b-zephyr-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: 61.88
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=wandb/mistral-7b-zephyr-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: 59.3
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=wandb/mistral-7b-zephyr-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: 78.53
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=wandb/mistral-7b-zephyr-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: 31.01
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=wandb/mistral-7b-zephyr-dpo
      name: Open LLM Leaderboard
---

[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/llm_surgery/gemma-zephyr)

# Mistral 7B Zephyr DPO V2

The [Zephyr](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) DPO recipe applied on top of Mistral 7B (new recipe with chatML format)

## Model description

- **Model type:** A 7.2B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.
- **Language(s) (NLP):** Primarily English
- **Finetuned from model:** [wandb/mistral-7b-zephyr-sft](https://huggingface.co/wandb/mistral-7b-zephyr-sft)

## Recipe

We trained using the [alignment handbook recipe](https://github.com/huggingface/alignment-handbook/blob/main/scripts/run_dpo.py) and logging to W&B

Visit the [W&B workspace here](https://wandb.ai/llm_surgery/gemma-zephyr?nw=nwusercapecape)

## Compute provided by Lambda Labs - 8xA100 80GB node

# [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_wandb__mistral-7b-zephyr-dpo)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |63.22|
|AI2 Reasoning Challenge (25-Shot)|63.05|
|HellaSwag (10-Shot)              |85.54|
|MMLU (5-Shot)                    |61.88|
|TruthfulQA (0-shot)              |59.30|
|Winogrande (5-shot)              |78.53|
|GSM8k (5-shot)                   |31.01|