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

    Neuronpedia

    INDEX
    Explanations

    words followed by slashes or parentheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    uniq
    0.67
    Botan
    0.67
    苜
    0.65
    聞
    0.64
     câte
    0.64
    ത്തിലേ
    0.63
    classic
    0.63
    possible
    0.61
    वेग
    0.61
    Flip
    0.61
    POSITIVE LOGITS
     Wasn
    0.67
     midwife
    0.65
     flound
    0.65
     amni
    0.64
     flounder
    0.63
     quorum
    0.63
     wasnt
    0.61
     formdata
    0.61
     apartments
    0.60
     éta
    0.59
    Act Density 0.054%

    No Known Activations