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

    Neuronpedia

    INDEX
    Explanations

    head

    np_max-act · gemini-2.0-flash

    This neuron fires on programming/configuration tokens—e.g. variable names, file paths, flags, and other code‐style strings in build scripts or Makefiles.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    inis
    -0.07
     SQL
    -0.07
    IBLE
    -0.07
     userId
    -0.07
    ât
    -0.07
     kidneys
    -0.07
    icularly
    -0.07
    -id
    -0.06
    υ
    -0.06
    ays
    -0.06
    POSITIVE LOGITS
    .loadtxt
    0.07
    ToWorld
    0.06
     EntityState
    0.06
    getDefault
    0.06
     Freeman
    0.06
    SHARE
    0.06
    pluck
    0.06
     způsob
    0.06
    .pathname
    0.05
    ForObject
    0.05
    Act Density 0.585%

    No Known Activations