README.md

metadata

language:
  - en
license: apache-2.0
tags:
  - merge
  - fine-tuned
datasets:
  - cognitivecomputations/dolphin
  - cognitivecomputations/dolphin-coder
  - ise-uiuc/Magicoder-OSS-Instruct-75K
  - teknium/openhermes
  - migtissera/Synthia-v1.3
base_model:
  - mistralai/Mistral-7B-Instruct-v0.2
  - ehartford/dolphin-2.2.1-mistral-7b
  - SciPhi/SciPhi-Mistral-7B-32k
  - ehartford/samantha-1.2-mistral-7b
  - Arc53/docsgpt-7b-mistral
  - HuggingFaceH4/zephyr-7b-beta
  - meta-math/MetaMath-Mistral-7B
  - Open-Orca/Mistral-7B-OpenOrca
  - openchat/openchat-3.5-1210
  - beowolx/MistralHermes-CodePro-7B-v1
  - TIGER-Lab/MAmmoTH-7B-Mistral
  - teknium/OpenHermes-2.5-Mistral-7B
  - Weyaxi/OpenHermes-2.5-neural-chat-v3-3-Slerp
  - mlabonne/NeuralHermes-2.5-Mistral-7B
model-index:
  - name: Mistral-7B-Merge-14-v0.3-ft-step-9984
    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: 62.54
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984
          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: 82.18
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984
          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: 62.92
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984
          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: 53.7
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984
          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: 75.61
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984
          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: 25.25
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=EmbeddedLLM/Mistral-7B-Merge-14-v0.3-ft-step-9984
          name: Open LLM Leaderboard

Model Description

This is fine-tuned model based on EmbeddedLLM/Mistral-7B-Merge-14-v0.3 for 9984 steps.

The dataset used are:

• dophin
• dolphin-coder
• Magicoder-OSS-Instruct-75K
• openhermes
• Synthia-v1.3

Chat Template

Prompt format: This model uses ChatML prompt format.

<|im_start|>system
You are Dolphin, a helpful AI assistant.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

Training

The model is scheduled to be fine-tuned for 3 epochs on 4 A100s using axolotl.

Shout-Out to OSS

Thank you to the Open Source AI community for bringing together marvelous code frameworks and datasets.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	60.37
AI2 Reasoning Challenge (25-Shot)	62.54
HellaSwag (10-Shot)	82.18
MMLU (5-Shot)	62.92
TruthfulQA (0-shot)	53.70
Winogrande (5-shot)	75.61
GSM8k (5-shot)	25.25