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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "t," particularly in negative or prohibitive contexts

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

    won't followed by a verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Dostupné
    -0.49
     InputDecoration
    -0.46
    SharedCtor
    -0.44
     Infór
    -0.44
    detectChanges
    -0.42
     méri
    -0.42
    حياته
    -0.42
    dhury
    -0.42
    солю
    -0.40
    hasErrors
    -0.40
    POSITIVE LOGITS
     wont
    0.89
     will
    0.74
    Will
    0.71
     Will
    0.70
    不会
    0.67
    不會
    0.67
    will
    0.66
     wireType
    0.64
     WILL
    0.64
    WILL
    0.60
    Act Density 0.006%

    No Known Activations