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

    Neuronpedia

    INDEX
    Explanations

    Live things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    them
    0.74
     THEM
    0.69
     них
    0.69
    posals
    0.68
     тях
    0.68
     them
    0.66
     courtship
    0.65
    െന്നാണ്
    0.65
     peas
    0.64
     loafers
    0.64
    POSITIVE LOGITS
     seemed
    1.06
     lasted
    1.01
     arose
    0.94
     flew
    0.93
     waited
    0.93
     sounded
    0.87
     occurred
    0.86
     went
    0.86
     appeared
    0.85
     functioned
    0.85
    Act Density 0.001%

    No Known Activations