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

    Neuronpedia

    INDEX
    Explanations

    letters in words probability

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     причиной
    -0.09
    �
    -0.08
     triggering
    -0.08
    団
    -0.08
     legends
    -0.08
     armored
    -0.08
    unwrap
    -0.08
     abgeschlossen
    -0.07
    RCT
    -0.07
    encher
    -0.07
    POSITIVE LOGITS
     अक्ष
    0.10
    (unique
    0.09
    _unique
    0.09
     UNIQUE
    0.09
     spelled
    0.09
     كلمة
    0.08
     بـ
    0.08
     duplicates
    0.08
     ['
    0.08
    unique
    0.08
    Act Density 0.012%

    No Known Activations