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

    Neuronpedia

    INDEX
    Explanations

    involvement and repetition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Quartz
    0.97
    口感
    0.95
    accuracy
    0.94
    👅
    0.94
     nonthermal
    0.93
     thumbnailUrl
    0.93
     pourrait
    0.90
    neon
    0.89
     expliquer
    0.89
     greenish
    0.88
    POSITIVE LOGITS
     multiple
    0.83
     multi
    0.82
     involvement
    0.80
    在一个
    0.75
     involved
    0.73
     compounded
    0.71
     multip
    0.71
     terlibat
    0.71
     repeatedly
    0.70
    團
    0.70
    Act Density 0.083%

    No Known Activations