Aura_v2_7B / README.md
jeiku's picture
Adding Evaluation Results (#1)
2c9d0e1 verified
metadata
language:
  - en
license: apache-2.0
library_name: transformers
base_model:
  - ResplendentAI/Paradigm_7B
  - jeiku/Theory_of_Mind_Mistral
  - ResplendentAI/Paradigm_7B
  - jeiku/selfbot_256_mistral
  - ResplendentAI/Paradigm_7B
  - jeiku/Gnosis_Reformatted_Mistral
  - ResplendentAI/Paradigm_7B
model-index:
  - name: Aura_v2_7B
    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: 73.46
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ResplendentAI/Aura_v2_7B
          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: 88.64
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ResplendentAI/Aura_v2_7B
          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: 63.97
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ResplendentAI/Aura_v2_7B
          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: 75.17
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ResplendentAI/Aura_v2_7B
          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: 84.45
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ResplendentAI/Aura_v2_7B
          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: 66.49
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ResplendentAI/Aura_v2_7B
          name: Open LLM Leaderboard

Aura v2

image/png

The second version of the Aura line is a direct improvement over the original. Expect poetic and eloquent outputs with real emotion behind them.

I recommend keeping the temperature around 1.5 or lower with a Min P value of 0.05. This model can get carried away with prose at higher temperature. I will say though that the prose of this model is distinct from the GPT 3.5/4 variant, and lends an air of humanity to the outputs. I am aware that this model is overfit, but that was the point of the entire exercise.

If you have trouble getting the model to follow an asterisks/quote format, I recommend asterisks/plaintext instead. This model skews toward shorter outputs, so be prepared to lengthen your introduction and examples if you want longer outputs.

This model responds best to ChatML for multiturn conversations.

This model, like all other Mistral based models, is compatible with a Mistral compatible mmproj file for multimodal vision capabilities in KoboldCPP.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 75.36
AI2 Reasoning Challenge (25-Shot) 73.46
HellaSwag (10-Shot) 88.64
MMLU (5-Shot) 63.97
TruthfulQA (0-shot) 75.17
Winogrande (5-shot) 84.45
GSM8k (5-shot) 66.49