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

    Neuronpedia

    INDEX
    Explanations

    yes/no

    np_max-act · gemini-2.0-flash

    This neuron responds to occurrences of the word “no,” especially when it appears in quotes to mark refusal or boundary-setting.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .SharedPreferences
    -0.07
    aln
    -0.07
    Shown
    -0.06
     очист
    -0.06
    .setToolTip
    -0.06
    ενο
    -0.06
    _mid
    -0.06
    HSV
    -0.06
     senin
    -0.06
    İL
    -0.06
    POSITIVE LOGITS
     prompts
    0.07
     supplementary
    0.07
     случ
    0.07
     vertically
    0.06
     nam
    0.06
     contract
    0.06
    ré
    0.06
    .lab
    0.06
     instantly
    0.06
     tuberculosis
    0.06
    Act Density 0.031%

    No Known Activations