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

    Neuronpedia

    INDEX
    Explanations

    I'd love to help

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    нице
    0.38
    无可
    0.37
     Presumably
    0.37
     دارید
    0.36
    明白
    0.35
    Pare
    0.35
     了
    0.34
    ிறார்கள்
    0.34
     والق
    0.34
     آمد
    0.33
    POSITIVE LOGITS
     rather
    0.78
    rather
    0.71
     Rather
    0.67
     gladly
    0.67
    Rather
    0.65
     piuttosto
    0.63
     plutôt
    0.61
     wager
    0.59
     honestly
    0.57
     eher
    0.57
    Act Density 0.009%

    No Known Activations