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

    Neuronpedia

    INDEX
    Explanations

    section numbering or phases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused1643>
    0.77
    <unused2156>
    0.76
    <unused2040>
    0.75
    <unused140>
    0.75
     UWGM
    0.75
    <unused2089>
    0.74
     pcMove
    0.74
    🤚
    0.74
     <-
    0.73
    ensureEqual
    0.72
    POSITIVE LOGITS
    1
    1.20
    2
    1.13
    3
    1.12
    6
    0.97
    II
    0.96
    5
    0.95
    III
    0.94
    4
    0.89
     Responsibility
    0.89
    8
    0.88
    Act Density 0.130%

    No Known Activations