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

    Neuronpedia

    INDEX
    Explanations

    common prefixes and their completions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ²,
    0.59
    witter
    0.59
     gimmick
    0.58
    ﺡ
    0.56
     gossip
    0.56
     spades
    0.55
    $,
    0.55
     flashbacks
    0.55
    }$,
    0.54
     carvings
    0.53
    POSITIVE LOGITS
     allgemeinen
    0.84
    ederal
    0.72
    ­
    0.72
    
    0.71
    Robert
    0.71
     Россий
    0.70
    及び
    0.70
    ‏
    0.70
    collegiate
    0.70
    䒨
    0.69
    Act Density 0.165%

    No Known Activations