[Eval Results] llama-3.2-1b-instruct on smalleval/mmlu-nano:mmlu_high_school_mathematics.jsonl - 20.8% Accuracy

[sauravpanda]

Evaluation Results

Metadata

• Dataset: smalleval/mmlu-nano:mmlu_high_school_mathematics.jsonl
• Model: llama-3.2-1b-instruct
• Evaluation Type: evaluation

Performance Metrics

• Accuracy: 20.8%
• Total Questions: 250
• Average Latency: 0.00ms
• Tokens/Second: 0.00
• Memory Usage: 884.14MB

System Information

• Browser: chrome 131
• OS: MacIntel
• CPU: 10 cores
• RAM: 8GB
• GPU: Not Available
• Timestamp: 2025-01-21 05:40:41

Evaluation Results Jan 21 2025 (3).json