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

    Neuronpedia

    INDEX
    Explanations

    how + intensity of feeling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    是要
    0.78
     >,
    0.78
     react
    0.73
    enge
    0.73
    是對
    0.73
    欷
    0.73
     이상의
    0.72
     reactivity
    0.71
     distinguishes
    0.70
    は
    0.70
    POSITIVE LOGITS
     incredible
    0.95
    incredible
    0.90
    amazing
    0.86
     increíble
    0.85
     incroyable
    0.84
    everyone
    0.83
    everything
    0.80
    Awesome
    0.79
    Amazing
    0.79
     amazing
    0.78
    Act Density 0.015%

    No Known Activations