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

    Neuronpedia

    INDEX
    Explanations

    HTML tags and code blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    та
    1.02
    ט
    0.75
    טת
    0.73
     préférable
    0.73
    ینی
    0.73
     testacé
    0.71
     assail
    0.71
    の前
    0.71
     unharmed
    0.71
    靔
    0.71
    POSITIVE LOGITS
    ig
    1.15
    i
    1.12
    et
    1.05
    ib
    1.00
    n
    1.00
    d
    0.94
    em
    0.89
    ou
    0.88
    or
    0.88
    elj
    0.84
    Act Density 0.032%

    No Known Activations