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

    Neuronpedia

    INDEX
    Explanations

    why questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mantenerse
    0.88
     escribir
    0.83
    Maintaining
    0.81
    \".
    0.80
    "#
    0.80
    maintain
    0.79
    refrac
    0.78
     enero
    0.75
    "..
    0.75
    nione
    0.75
    POSITIVE LOGITS
     compels
    0.70
     commands
    0.69
     commitment
    0.68
     schema
    0.68
     VPN
    0.68
    سون
    0.66
     เรียก
    0.65
    де
    0.64
     matching
    0.64
     referral
    0.64
    Act Density 0.004%

    No Known Activations