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

    Neuronpedia

    INDEX
    Explanations

    break down, pick up

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﻣ
    0.96
    俶
    0.80
    StartsWith
    0.78
    ຫມ
    0.76
    næ
    0.76
    ຫລ
    0.75
    lbrace
    0.74
    ArrivalTime
    0.74
     núcle
    0.73
    ﺣ
    0.73
    POSITIVE LOGITS
     up
    2.68
     Up
    1.61
    up
    1.60
     down
    1.44
     upas
    1.44
    Up
    1.40
     upp
    1.29
     UP
    1.28
     out
    1.28
     lên
    1.17
    Act Density 0.479%

    No Known Activations