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

    Neuronpedia

    INDEX
    Explanations

    there or it followed by 's

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     chooses
    0.53
     quits
    0.50
     selects
    0.47
     हुन्छ
    0.47
     Calculates
    0.47
     avoids
    0.47
     succeeds
    0.46
     attends
    0.45
     learns
    0.45
    把
    0.44
    POSITIVE LOGITS
     originated
    0.95
     underwent
    0.81
     arose
    0.78
     came
    0.77
     evolved
    0.74
     sprang
    0.74
     began
    0.73
     appeared
    0.72
     occurred
    0.71
     debuted
    0.71
    Act Density 0.628%

    No Known Activations