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

    Neuronpedia

    INDEX
    Explanations

    legal documents

    np_max-act · gemini-2.0-flash

    first-person declarative phrases where the speaker asserts or performs an action (e.g., “I” followed by a verb indicating intent, action, or attestation), often in formal or persuasive contexts.

    oai_token-act-pair · gpt-5Triggered by @vetterc0

    first-person singular statements or declarations, typically with “I” plus a verb expressing action, intent, or authorship.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Bloss
    -0.08
     she
    -0.07
     he
    -0.07
     pre
    -0.07
    讥
    -0.07
    !!
    -0.07
     extracting
    -0.07
     forfe
    -0.07
    גר
    -0.07
     para
    -0.07
    POSITIVE LOGITS
    ’es
    0.08
    	pid
    0.08
    usaha
    0.07
    ilent
    0.07
    仔细
    0.07
     Common
    0.07
     Tình
    0.07
    켠
    0.06
    usu
    0.06
    	ui
    0.06
    Act Density 2.336%

    No Known Activations