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

    Neuronpedia

    INDEX
    Explanations

    naming and titling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Textbook
    0.77
     textbook
    0.77
     instrucción
    0.70
     routine
    0.70
     pengendalian
    0.70
     bounded
    0.68
     textbooks
    0.68
    ỉnh
    0.68
     Instruction
    0.67
     najve
    0.67
    POSITIVE LOGITS
     catchy
    1.55
     puns
    1.32
     syllables
    1.22
     evocative
    1.20
     connotations
    1.16
     pronounce
    1.10
    Naming
    1.08
     conveys
    1.07
     названия
    1.07
     names
    1.07
    Act Density 0.583%

    No Known Activations