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

    Neuronpedia

    INDEX
    Explanations

    how to instruct

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     인한
    0.56
     edging
    0.55
    ผู้
    0.53
     Remarkably
    0.52
    ร่วม
    0.49
     surprisingly
    0.48
    च्या
    0.48
    ਪ
    0.47
    تان
    0.46
     bowing
    0.46
    POSITIVE LOGITS
     that
    0.91
     what
    0.89
     что
    0.85
     që
    0.84
    that
    0.82
    what
    0.82
     दैट
    0.81
     दट
    0.75
     أنه
    0.74
     how
    0.74
    Act Density 0.012%

    No Known Activations