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

    Neuronpedia

    INDEX
    Explanations

    fields

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Mathemat
    0.63
    Rock
    0.61
    твор
    0.61
    اردوش
    0.61
     يصير
    0.59
     Overhead
    0.59
    pannt
    0.58
     Rock
    0.57
     Hygien
    0.57
    Fragmentation
    0.56
    POSITIVE LOGITS
     field
    3.06
     fields
    2.98
     Field
    2.63
     campo
    2.60
    fields
    2.58
    Field
    2.57
     campos
    2.46
    field
    2.46
     Fields
    2.45
    Fields
    2.37
    Act Density 0.428%

    No Known Activations