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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks that indicate dialogue or quotation

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

    start of quote or code

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

    It detects the start of a quoted utterance (opening quotation marks / the beginning of a dialogue turn).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    Personendaten
    -0.74
     Autorizaciones
    -0.66
     resourceCulture
    -0.64
    PostInfinity
    -0.64
    ">//
    -0.62
     صوتيه
    -0.62
     चीज़ों
    -0.56
    OGND
    -0.55
    OCCURRED
    -0.55
     كومونز
    -0.55
    POSITIVE LOGITS
    GORITH
    0.35
     Efficient
    0.34
     Initially
    0.33
     Schatz
    0.33
     butt
    0.33
     initially
    0.32
    arache
    0.32
     Right
    0.32
     re
    0.31
     proceeds
    0.31
    Act Density 0.099%

    No Known Activations