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

    Neuronpedia

    INDEX
    Explanations

    know future number

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     rå
    0.43
     Chilean
    0.43
    पछि
    0.42
    on
    0.39
     کہنے
    0.39
    ergic
    0.38
    }\{\
    0.37
    Chile
    0.37
     [,
    0.37
     וח
    0.37
    POSITIVE LOGITS
     attribution
    0.44
     accretion
    0.43
     assume
    0.39
     envelope
    0.39
     accrued
    0.39
    خراج
    0.39
    某个
    0.38
     accru
    0.38
     determinados
    0.38
    瀉
    0.38
    Act Density 0.000%

    No Known Activations