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

    Neuronpedia

    INDEX
    Explanations

    swim through water/turbine/class/from jeans/openings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    タ
    0.50
    лі
    0.49
     lifts
    0.47
    دست
    0.46
     lifted
    0.46
    日
    0.46
    ুই
    0.45
     fluttering
    0.44
    俊
    0.44
    ست
    0.43
    POSITIVE LOGITS
     راه
    0.47
     Swallow
    0.47
    飡
    0.47
     आठवीं
    0.45
    MANY
    0.45
    IO
    0.43
     รัก
    0.43
    etable
    0.43
     سواء
    0.43
    atsion
    0.43
    Act Density 0.000%

    No Known Activations