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

    Neuronpedia

    INDEX
    Explanations

    conversational phrases and the dynamics of dialogue

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

    reporting speech or requirement

    np_acts-logits-general · gemini-2.5-flash-lite

    spots words and phrases that indicate reported speech, claims, or that someone said/thought something (reporting/hearsay language).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    mità
    -0.37
     Freuden
    -0.35
     Réponses
    -0.35
    olingo
    -0.34
    ofday
    -0.34
     boste
    -0.34
     intptr
    -0.34
     Ludlow
    -0.33
     Arca
    -0.33
     Hamlet
    -0.33
    POSITIVE LOGITS
    UseVisualStyle
    0.51
    ]='\
    0.50
    OpenHelper
    0.46
    featureID
    0.46
     nakalista
    0.45
    transQ
    0.45
    %)$
    0.43
    urlpatterns
    0.42
    فاده
    0.42
    ****/
    0.42
    Act Density 0.324%

    No Known Activations