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

    Neuronpedia

    INDEX
    Explanations

    timestamps and numerical identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    瀉
    0.45
    Shore
    0.45
    泻
    0.45
    氙
    0.45
    Acetyl
    0.45
    饴
    0.42
     flange
    0.41
     tannin
    0.41
    鉭
    0.41
     canalicul
    0.41
    POSITIVE LOGITS
    XXXXXXXX
    0.53
    6
    0.52
    xxxxxxxx
    0.51
    8
    0.49
     billion
    0.47
    example
    0.46
    7
    0.46
    1
    0.46
    9
    0.44
    …
    0.43
    Act Density 0.010%

    No Known Activations