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

    Neuronpedia

    INDEX
    Explanations

    compression

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     compression
    2.91
     compress
    2.83
    压缩
    2.80
     Compression
    2.75
     compressed
    2.73
    Compression
    2.70
     compressing
    2.66
    Compress
    2.61
    壓縮
    2.58
    compression
    2.56
    POSITIVE LOGITS
    arante
    0.39
     Rud
    0.39
     খালে
    0.39
     hexagonal
    0.37
     Brexit
    0.37
    ត្តិ
    0.36
    stagram
    0.35
    ScheduledAction
    0.35
    皑
    0.35
    嗗
    0.35
    Act Density 0.043%

    No Known Activations