KnutJaegersberg's picture
Adding Evaluation Results (#1)
e6810f6 verified
metadata
license: other
datasets:
  - totally-not-an-llm/EverythingLM-data-V3
license_name: yi-license
license_link: LICENSE
pipeline_tag: text-generation
model-index:
  - name: Deacon-34b-qlora-adapter
    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: 64.85
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=KnutJaegersberg/Deacon-34b-qlora-adapter
          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: 85.56
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=KnutJaegersberg/Deacon-34b-qlora-adapter
          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: 76.38
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=KnutJaegersberg/Deacon-34b-qlora-adapter
          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: 56.21
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=KnutJaegersberg/Deacon-34b-qlora-adapter
          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: 83.11
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=KnutJaegersberg/Deacon-34b-qlora-adapter
          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: 62.24
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=KnutJaegersberg/Deacon-34b-qlora-adapter
          name: Open LLM Leaderboard

image/png

The perfect organism.

An adapter for KnutJaegersberg/Yi-34B-Llamafied. 5 epochs with NEFTune.

Prompt Example:

### System:

You are an AI assistant. User will give you a task. Your goal is to complete the task as faithfully as you can. While performing the task think step-by-step and justify your steps.

### Instruction: 

How do you fine tune a large language model? 

### Response:

image/png

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 71.39
AI2 Reasoning Challenge (25-Shot) 64.85
HellaSwag (10-Shot) 85.56
MMLU (5-Shot) 76.38
TruthfulQA (0-shot) 56.21
Winogrande (5-shot) 83.11
GSM8k (5-shot) 62.24