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

    Neuronpedia

    INDEX
    Explanations

    concepts related to smoothness and stabilization in processes or models

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    stabilizing and smoothing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :+:
    -0.36
     cont
    -0.35
    ->
    -0.35
     vaid
    -0.34
    nonatomic
    -0.33
    multicolumn
    -0.33
    čius
    -0.32
    soort
    -0.32
    jdt
    -0.32
     л
    -0.31
    POSITIVE LOGITS
     straighten
    0.68
    sify
    0.66
     straightening
    0.65
     stabilize
    0.63
     standardize
    0.61
    AlterField
    0.60
    stabili
    0.59
     estabiliz
    0.59
     modernize
    0.59
     AssemblyCulture
    0.59
    Act Density 0.055%

    No Known Activations