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

    Neuronpedia

    INDEX
    Explanations

    BigQuery SQL queries and tables

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    惭
    0.38
    ოვნ
    0.37
     দেড়
    0.33
    envs
    0.33
     binder
    0.32
    buildings
    0.32
    igit
    0.32
     থাকতে
    0.32
    BITS
    0.32
     expanding
    0.31
    POSITIVE LOGITS
     кристал
    0.39
     cristal
    0.38
     Label
    0.38
    cript
    0.37
     Error
    0.35
     guardian
    0.35
     Id
    0.34
    Loaded
    0.34
    Label
    0.34
     Cancel
    0.34
    Act Density 0.001%

    No Known Activations