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

    Neuronpedia

    INDEX
    Explanations

    speech fillers

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     др
    1.02
     ("
    0.98
     असल्यास
    0.96
    므로
    0.95
     এরূপ
    0.93
     אך
    0.91
     이는
    0.91
     "
    0.90
     (\"
    0.88
     ancak
    0.84
    POSITIVE LOGITS
    …”
    1.17
     really
    1.16
    っていう
    1.15
    …’
    1.10
    …
    1.08
     ähm
    1.07
     [
    1.06
     sort
    1.06
     yeah
    1.05
    ,”
    1.03
    Act Density 0.330%

    No Known Activations