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

    Neuronpedia

    INDEX
    Explanations

    the leading word of events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    意思是
    0.38
     topologies
    0.38
    ategor
    0.38
     discontinuities
    0.37
    这个人
    0.36
     paradigms
    0.36
     subsections
    0.35
    自体
    0.34
     instantiated
    0.34
     sopra
    0.34
    POSITIVE LOGITS
     শহরের
    0.47
     newfound
    0.43
     lucrative
    0.43
     pricey
    0.43
     duo
    0.42
    公司的
    0.40
     ordeal
    0.40
     fledgling
    0.40
     makeshift
    0.39
     компанию
    0.39
    Act Density 0.005%

    No Known Activations