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

    Neuronpedia

    INDEX
    Explanations

    grip tape, patch cable, ball presses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Schema
    0.55
    Ordinal
    0.55
    Semantic
    0.54
    Daemon
    0.54
    Uppercase
    0.53
     Schema
    0.52
     semantic
    0.52
    Translations
    0.51
    Preprocessing
    0.50
    Immutable
    0.50
    POSITIVE LOGITS
    less
    0.80
     style
    0.77
    style
    0.76
    式の
    0.72
    式
    0.70
    board
    0.68
    proof
    0.66
    タイプの
    0.64
     type
    0.64
     tipi
    0.63
    Act Density 0.571%

    No Known Activations