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

    Neuronpedia

    INDEX
    Explanations

    responding to requests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     yummy
    0.53
     goodies
    0.52
     nifty
    0.46
     nice
    0.45
     eureka
    0.45
     snag
    0.44
     handy
    0.44
     Enh
    0.44
     Blessing
    0.44
     얘
    0.43
    POSITIVE LOGITS
     hypothetical
    0.87
     requested
    0.86
     hypot
    0.77
    requested
    0.75
     Hypot
    0.74
     Requested
    0.69
    Requested
    0.68
     requesting
    0.66
     attempting
    0.62
     satire
    0.61
    Act Density 0.357%

    No Known Activations