Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    logging handlers and formatters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Analytics
    0.45
    統計
    0.43
    statistical
    0.43
    ීර
    0.42
     statist
    0.42
    ସ
    0.41
    প্রশ্ন
    0.41
    raphene
    0.41
     analytics
    0.40
    Statistical
    0.40
    POSITIVE LOGITS
     handlers
    0.95
     formatter
    0.85
    handlers
    0.84
     handler
    0.81
    Formatter
    0.81
     Handler
    0.77
    formatter
    0.75
    Handler
    0.68
    Handlers
    0.66
    handler
    0.65
    Act Density 0.007%

    No Known Activations