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

    Neuronpedia

    INDEX
    Explanations

    word counts for spoken text

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     legitimacy
    -0.08
    .Texture
    -0.08
     tij
    -0.08
    _geometry
    -0.08
    tails
    -0.07
    \File
    -0.07
     herb
    -0.07
    efile
    -0.07
     absor
    -0.07
     Datei
    -0.07
    POSITIVE LOGITS
     paced
    0.12
    速度
    0.11
     spoken
    0.11
    -paced
    0.11
     pace
    0.11
     বক্ত
    0.11
    Speech
    0.10
     speech
    0.10
     Speech
    0.10
     slowed
    0.10
    Act Density 0.004%

    No Known Activations