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

    Neuronpedia

    INDEX
    Explanations

    adverbs of certainty or manner

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ruining
    0.51
     ruins
    0.50
     hates
    0.49
     screwed
    0.49
     ruined
    0.48
     hurting
    0.47
     messed
    0.47
     messing
    0.47
     suka
    0.46
     usan
    0.46
    POSITIVE LOGITS
     undeniably
    0.54
     inevitably
    0.54
     arguably
    0.54
     invariably
    0.51
    argu
    0.47
     undoubtedly
    0.46
     ultimately
    0.46
     inadvertently
    0.45
     inescap
    0.45
     subtly
    0.43
    Act Density 0.265%

    No Known Activations