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

    Neuronpedia

    INDEX
    Explanations

    Legal disclaimers

    np_max-act · gemini-2.0-flash

    The neuron detects legal fault‐and‐liability terms—especially words like “misconduct,” “negligence,” “intentional,” or “fraudulent” that signal wrongdoing or liability.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     linger
    -0.06
     submar
    -0.06
    .TextAlignment
    -0.06
    	tile
    -0.06
     lifecycle
    -0.06
    Increasing
    -0.06
    signal
    -0.06
    tr
    -0.06
     mock
    -0.06
    Tiles
    -0.06
    POSITIVE LOGITS
    resas
    0.08
     Güney
    0.07
     dot
    0.07
    ignty
    0.06
     Dot
    0.06
    rox
    0.06
    exc
    0.06
     confidential
    0.06
     meslek
    0.06
    emploi
    0.06
    Act Density 0.004%

    No Known Activations