README.md

metadata

license: apache-2.0
model-index:
  - name: openllama-7b-icl
    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: 47.95
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=itsliupeng/openllama-7b-icl
          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: 77.04
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=itsliupeng/openllama-7b-icl
          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: 44.37
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=itsliupeng/openllama-7b-icl
          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: 37.06
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=itsliupeng/openllama-7b-icl
          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: 70.17
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=itsliupeng/openllama-7b-icl
          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: 10.99
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=itsliupeng/openllama-7b-icl
          name: Open LLM Leaderboard

Train openllama-7b with in-context leanrning

A Reproduction of OpenLLaMA using 128 H100 GPUs in Bfloat16.

The pretrain data consists of Falcon, Starcoder, and the wikipedia, arxiv, books, stackexchange from RedPajama. In total, this encompassed nearly 1 trillion tokens.

The model was trained over a single epoch, incorporating 2000 warm-up steps and a cosine learning rate schedule, starting at 3e-5 with 4M batch size.

The sole distinction from the OpenLLaMA 7B Base lies in the organization of Falcon documents, which follows the methodology outlined in this arXiv paper.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	47.93
AI2 Reasoning Challenge (25-Shot)	47.95
HellaSwag (10-Shot)	77.04
MMLU (5-Shot)	44.37
TruthfulQA (0-shot)	37.06
Winogrande (5-shot)	70.17
GSM8k (5-shot)	10.99