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

    Neuronpedia

    INDEX
    Explanations

    COPY command in Dockerfiles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.70
     or
    0.65
    D
    0.58
    त
    0.58
    ."
    0.57
    ه
    0.57
    ات
    0.55
     
    0.52
    ↵↵
    0.52
    .*
    0.52
    POSITIVE LOGITS
     знания
    0.53
    fireFlower
    0.50
    掟
    0.50
     filede
    0.50
    昛
    0.49
    骢
    0.49
     χαρακτη
    0.49
     снижение
    0.49
    吟
    0.49
    fia
    0.48
    Act Density 0.000%

    No Known Activations