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

    Neuronpedia

    INDEX
    Explanations

    technical concepts and structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    \_
    1.04
    _
    1.03
    Nums
    0.94
    '=
    0.93
    _:
    0.88
     buffs
    0.86
    Cols
    0.86
    oS
    0.85
    ToPointer
    0.84
     Esther
    0.84
    POSITIVE LOGITS
     how
    0.78
     comprehensive
    0.62
    oweit
    0.61
     …
    0.60
     good
    0.59
    how
    0.59
     knowing
    0.58
     simple
    0.58
    estimating
    0.58
     perfect
    0.57
    Act Density 0.119%

    No Known Activations