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

    Neuronpedia

    INDEX
    Explanations

    error handling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    idbody
    0.73
     rebates
    0.68
    𓈒
    0.64
     teint
    0.64
    itchie
    0.64
    ölf
    0.62
    鹈
    0.62
    ⌵
    0.62
    🕶
    0.61
    radian
    0.60
    POSITIVE LOGITS
     error
    1.78
     Error
    1.76
    error
    1.69
    Error
    1.69
     errors
    1.54
    エラー
    1.54
    Exception
    1.49
     Exception
    1.44
     Errors
    1.43
     errores
    1.42
    Act Density 0.415%

    No Known Activations