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

    Neuronpedia

    INDEX
    Explanations

    instances of reprimands or corrections in a text

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

    scold, rebuke, admonish, censure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     createSlice
    -0.60
    UrlResolution
    -0.59
    Spoljašnje
    -0.53
     surla
    -0.47
    wapV
    -0.46
    PreferredItem
    -0.44
    ########.
    -0.42
    numerusform
    -0.42
     Meksiku
    -0.42
    UserScript
    -0.41
    POSITIVE LOGITS
     scold
    1.07
     admon
    1.01
     rebuke
    0.99
     scolded
    0.99
     admonition
    0.85
     rebu
    0.78
     chasti
    0.78
     repri
    0.76
     reproach
    0.73
     censure
    0.69
    Act Density 0.029%

    No Known Activations