README.md

metadata

language:
  - en
license: llama3
tags:
  - Llama-3
  - instruct
  - finetune
  - chatml
  - gpt4
  - synthetic data
  - distillation
  - function calling
  - json mode
  - axolotl
  - roleplaying
  - chat
base_model: NousResearch/Hermes-3-Llama-3.2-3B
widget:
  - example_title: Hermes 3
    messages:
      - role: system
        content: >-
          You are a sentient, superintelligent artificial general intelligence,
          here to teach and assist me.
      - role: user
        content: >-
          Write a short story about Goku discovering kirby has teamed up with
          Majin Buu to destroy the world.
model-index:
  - name: mrcuddle/Dark-Hermes3-Llama3.2-3B
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          type: lambada_openai
          name: LAMBADA OpenAI
          config: default
          split: test
        metrics:
          - type: accuracy
            value: 0.6837
            name: Accuracy
            config: none
            args:
              n-shot: 0
            stderr: 0.0065
          - type: perplexity
            value: 3.7577
            name: Perplexity
            config: none
            args:
              n-shot: 0
            stderr: 0.0933
library_name: transformers

Model Card

"Dark-Hermes3-Llama3.2-3B" is a fine-tuned version of NousResearch's Hermes 3B.

Training Details

Base Model:

• Hermes 3B by NousResearch

Fine-Tuning Datasets:

• Synthetic-Dark-RP
• Luminous_Opus
• Synthetic-RP

Tools Used:

• AutoTrain
• Axolotl