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

    Neuronpedia

    INDEX
    Explanations

    opening quotation marks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fre
    0.67
    ，《
    0.66
     supervision
    0.65
     stimul
    0.65
     comen
    0.64
    -
    0.64
    ,
    0.63
     bols
    0.63
    よい
    0.61
    itters
    0.61
    POSITIVE LOGITS
    They
    0.94
     éton
    0.92
    Warning
    0.89
    与其
    0.87
    Their
    0.87
    When
    0.86
    Goodbye
    0.86
    Don
    0.86
    𝘞
    0.86
    Certain
    0.86
    Act Density 0.194%

    No Known Activations