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

    Neuronpedia

    INDEX
    Explanations

    AI search and state expansion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    }^{(\
    -0.88
    möglichen
    -0.75
    chaik
    -0.74
    中に
    -0.73
     Pastoral
    -0.71
    hadas
    -0.71
     surveyor
    -0.69
     olden
    -0.69
     Picking
    -0.69
    ^{(
    -0.68
    POSITIVE LOGITS
    Puzzle
    1.26
     puzzle
    1.17
     Puzzle
    1.09
    puzzle
    1.07
     goal
    1.02
    aima
    0.98
    Expand
    0.98
     successors
    0.97
     createState
    0.95
     expanded
    0.91
    Act Density 0.009%

    No Known Activations