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

    Neuronpedia

    INDEX
    Explanations

    research publications

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     pseudo
    -0.07
     awaken
    -0.07
     mw
    -0.07
     proprietor
    -0.07
    消失
    -0.06
    멍
    -0.06
     marketers
    -0.06
     detrimental
    -0.06
    ukkit
    -0.06
     isConnected
    -0.06
    POSITIVE LOGITS
    annah
    0.07
    	service
    0.07
    娴
    0.07
    RA
    0.06
    bl
    0.06
     Gäste
    0.06
     conce
    0.06
    fe
    0.06
    .site
    0.06
    (tableView
    0.06
    Act Density 0.005%

    No Known Activations