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

    Neuronpedia

    INDEX
    Explanations

    basically, mainly, explanation follows

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     perempt
    0.90
     auspices
    0.90
     unequivocal
    0.84
     appreciably
    0.82
     bona
    0.81
     willkommen
    0.80
    genuine
    0.78
    preceq
    0.78
    ٭
    0.78
     sumptuous
    0.78
    POSITIVE LOGITS
     basically
    1.52
    想要
    1.50
    非常的
    1.48
    很多的
    1.39
    basically
    1.37
     Basically
    1.34
    Basically
    1.34
    有很多
    1.33
     মূলত
    1.31
     ponieważ
    1.30
    Act Density 0.151%

    No Known Activations