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

    Neuronpedia

    INDEX
    Explanations

    keyboard commands

    np_max-act · gemini-2.0-flash

    This neuron fires most strongly on subword tokens ending in “ing,” i.e. the gerund/progressive “-ing” suffix.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     resisted
    -0.07
    .file
    -0.07
    pe
    -0.06
    쟁
    -0.06
    .exam
    -0.06
    Bar
    -0.06
    ротив
    -0.06
    ioneer
    -0.06
     axes
    -0.06
     melted
    -0.06
    POSITIVE LOGITS
    serir
    0.08
    。（
    0.07
    (LogLevel
    0.07
    �
    0.07
     vyh
    0.07
     граду
    0.07
    Chr
    0.06
    �
    0.06
    审
    0.06
     nouve
    0.06
    Act Density 0.027%

    No Known Activations