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

    Neuronpedia

    INDEX
    Explanations

    parentheses or certain phrases after specific tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﻪ
    0.53
    gpt
    0.50
    ົງ
    0.48
    OVER
    0.48
    EVA
    0.45
    べる
    0.45
    żyć
    0.45
    the
    0.44
    жнему
    0.44
    యోజ
    0.44
    POSITIVE LOGITS
     Arrays
    0.54
     arrays
    0.52
     näch
    0.52
    ್ಟ
    0.49
     salads
    0.48
     Sle
    0.47
    หล
    0.47
     workstations
    0.47
     LU
    0.47
     Lan
    0.47
    Act Density 0.000%

    No Known Activations