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

    Neuronpedia

    INDEX
    Explanations

    Code/numerical data

    np_max-act · gemini-2.0-flash

    lines or tokens that look like code, logs, stack traces, or other programming/technical text (identifiers, timestamps, punctuation-heavy tokens).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    .fig
    -0.06
    \a
    -0.06
    	onClick
    -0.06
    	f
    -0.06
     ho
    -0.06
     welt
    -0.05
     trä
    -0.05
    .DateTimePicker
    -0.05
     sleeve
    -0.05
    _extract
    -0.05
    POSITIVE LOGITS
     Uint
    0.07
     evidently
    0.06
     GTA
    0.06
     pdata
    0.06
     Newsp
    0.06
     Lionel
    0.06
     Companies
    0.06
    eterminate
    0.06
     Turkish
    0.06
     αυ
    0.06
    Act Density 0.320%

    No Known Activations