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

    Neuronpedia

    INDEX
    Explanations

    initialization and setup

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     xd
    -0.07
    WORK
    -0.07
     merges
    -0.06
     outlines
    -0.06
     Pasadena
    -0.06
     MediaType
    -0.06
     '<?
    -0.06
     priests
    -0.06
     UIG
    -0.06
    )?;↵
    -0.06
    POSITIVE LOGITS
     вещ
    0.07
     Squ
    0.06
     "('
    0.06
    (loss
    0.06
    ôle
    0.06
    proc
    0.06
     throughput
    0.06
    Dice
    0.06
     jur
    0.06
    Johnny
    0.06
    Act Density 0.020%

    No Known Activations