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

    Neuronpedia

    INDEX
    Explanations

    chain of thought

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     floppy
    0.27
     erg
    0.27
     arr
    0.26
     eff
    0.26
     initializes
    0.26
     veget
    0.25
     Frey
    0.25
     jitter
    0.25
     tracers
    0.25
     distr
    0.25
    POSITIVE LOGITS
    産業
    0.29
    Су
    0.29
    買取
    0.29
    Cit
    0.28
    Ukraine
    0.27
     ойношот
    0.27
     இதே
    0.27
    anamo
    0.26
    а
    0.26
    Perhaps
    0.25
    Act Density 0.001%

    No Known Activations