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

    Neuronpedia

    INDEX
    Explanations

    logging and errors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    തിരെ
    0.67
    .,
    0.67
     against
    0.67
    ｗｗ
    0.66
    ,),
    0.66
     cures
    0.65
     waivers
    0.64
     indexes
    0.64
     villains
    0.64
    ,[
    0.63
    POSITIVE LOGITS
     "¿
    1.26
    Text
    1.08
     Phrases
    1.02
     Text
    1.02
    textContent
    1.02
    ("
    1.02
     Phrase
    0.98
    テキスト
    0.98
    Phrase
    0.97
     phrase
    0.95
    Act Density 0.546%

    No Known Activations