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

    Neuronpedia

    INDEX
    Explanations

    watches

    np_max-act-logits · gemini-2.0-flashTriggered by @majianfei030706
    New Auto-Interp
    Negative Logits
     Equ
    -0.10
     communicating
    -0.07
     Eph
    -0.07
     domestically
    -0.07
    见面
    -0.07
     technicians
    -0.07
     reacting
    -0.07
    هى
    -0.07
     Iraq
    -0.07
     adjusting
    -0.07
    POSITIVE LOGITS
    .FONT
    0.07
     Lara
    0.07
    READ
    0.06
    Prompt
    0.06
    	inst
    0.06
    .Speed
    0.06
    .INTER
    0.06
    ヘル
    0.06
    :"<<
    0.06
    CHOOL
    0.06
    Act Density 0.045%

    No Known Activations