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

    Neuronpedia

    INDEX
    Explanations

    haiku requests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Tash
    0.76
     opts
    0.75
    blk
    0.75
     Pearls
    0.70
     جیس
    0.70
    <unused746>
    0.70
    ୬
    0.70
    ਾਰ
    0.69
    <unused398>
    0.68
    ุล
    0.67
    POSITIVE LOGITS
     brother
    0.72
     bat
    0.66
     brothers
    0.65
     sibling
    0.64
     overnight
    0.64
     cutting
    0.64
     Bat
    0.63
    屣
    0.62
     भाइयों
    0.62
     themselves
    0.61
    Act Density 0.006%

    No Known Activations