README.md

metadata

license: llama3.2
base_model:
  - meta-llama/Llama-3.2-1B-Instruct
model-index:
  - name: Llama-3.2-SUN-2.4B-v1.0.0
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 56.37
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=meditsolutions/Llama-3.2-SUN-2.4B-v1.0.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 7.21
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=meditsolutions/Llama-3.2-SUN-2.4B-v1.0.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 4.83
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=meditsolutions/Llama-3.2-SUN-2.4B-v1.0.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 1.01
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=meditsolutions/Llama-3.2-SUN-2.4B-v1.0.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 3.02
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=meditsolutions/Llama-3.2-SUN-2.4B-v1.0.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 6.03
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=meditsolutions/Llama-3.2-SUN-2.4B-v1.0.0
          name: Open LLM Leaderboard
datasets:
  - argilla/OpenHermesPreferences
  - argilla/magpie-ultra-v0.1
  - argilla/Capybara-Preferences-Filtered
  - mlabonne/open-perfectblend
  - HuggingFaceTB/everyday-conversations-llama3.1-2k
  - WizardLMTeam/WizardLM_evol_instruct_V2_196k
  - ProlificAI/social-reasoning-rlhf
language:
  - en

MedIT SUN 2.5B

Base Model

• Llama 3.2 1B

Extended Size

• 1B to 2.5B parameters

Extension Method

• Proprietary technique developed by MedIT Solutions

Fine-tuning

• Open (or open subsets allowing for commercial use) open datasets from HF
• Open (or open subsets allowing for commercial use) SFT datasets from HF

Training Status

• Current version: chat-1.0.0

Key Features

• Built on Llama 3.2 architecture
• Expanded from 1B to 2.47B parameters
• Optimized for open-ended conversations
• Incorporates supervised fine-tuning for improved performance

Use Case

• General conversation and task-oriented interactions

Limitations As the model is still in training, performance and capabilities may vary. Users should be aware that the model is not in its final form and may exhibit inconsistencies or limitations typical of in-progress AI models.

Disclaimer and Safety Considerations The Model is designed to be used as a smart assistant but not as a knowledge source within your applications, systems, or environments. It is not intended to provide 100% accurate answers, especially in scenarios where high precision and accuracy are crucial.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	13.08
IFEval (0-Shot)	56.37
BBH (3-Shot)	7.21
MATH Lvl 5 (4-Shot)	4.83
GPQA (0-shot)	1.01
MuSR (0-shot)	3.02
MMLU-PRO (5-shot)	6.03