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

    Neuronpedia

    INDEX
    Explanations

    proper nouns followed by words like said or declared

    oai_token-act-pair · gemini-2.0-flash

    words related to speech

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    dero
    -0.54
     }.
    -0.54
     grand
    -0.53
    artifactId
    -0.48
    ándola
    -0.47
    ########.
    -0.47
    zean
    -0.47
     binary
    -0.46
     wunderbar
    -0.46
    Bibliograf
    -0.46
    POSITIVE LOGITS
     said
    0.66
    Посилання
    0.63
     told
    0.62
    contentValues
    0.60
     explained
    0.58
     argued
    0.57
    hints
    0.57
    CopyWith
    0.56
    UnusedPrivate
    0.56
    iprot
    0.54
    Act Density 0.734%

    No Known Activations