README.md

metadata

license: llama3.2
model-index:
  - name: Llama-3.2-1B-Instruct-CrashCourse12K
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: wis-k/instruction-following-eval
          split: train
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 53.95
            name: averaged accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=agentlans%2FLlama-3.2-1B-Instruct-CrashCourse12K
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: SaylorTwift/bbh
          split: test
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 9.39
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=agentlans%2FLlama-3.2-1B-Instruct-CrashCourse12K
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: lighteval/MATH-Hard
          split: test
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 6.57
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=agentlans%2FLlama-3.2-1B-Instruct-CrashCourse12K
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          split: train
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 0
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=agentlans%2FLlama-3.2-1B-Instruct-CrashCourse12K
          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: 1.2
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=agentlans%2FLlama-3.2-1B-Instruct-CrashCourse12K
          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: 8.99
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/?search=agentlans%2FLlama-3.2-1B-Instruct-CrashCourse12K
          name: Open LLM Leaderboard

Model Card: agentlans/Llama-3.2-1B-Instruct-CrashCourse12K

Model Overview

• Base Model: Llama-3.2-1B-Instruct
• Fine-tuning Type: Supervised Fine-Tuning (SFT)
• Dataset: agentlans/crash-course (12,000 rows)
• Purpose: Enhanced instruction-following capabilities

Training Details

• Method: Supervised fine-tuning on high-quality instruction dataset
• Training Rows: 12,000
• Objective: Improve task completion and instruction understanding

Performance

• Optimized for multi-task instruction following
• Improved zero-shot and few-shot performance
• Enhanced reasoning and response coherence

Limitations

• 1B parameter model with constrained complex reasoning
• Knowledge cutoff: December 2023
• Potential inherited biases from base model and training data

Recommended Use

• General instruction-based tasks
• Educational content generation
• Simple reasoning and task completion

Open LLM Leaderboard Evaluation Results

Detailed results can be found here! Summarized results can be found here!

Metric	Value (%)
Average	13.35
IFEval (0-Shot)	53.95
BBH (3-Shot)	9.39
MATH Lvl 5 (4-Shot)	6.57
GPQA (0-shot)	0.00
MuSR (0-shot)	1.20
MMLU-PRO (5-shot)	8.99