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

    Neuronpedia

    INDEX
    Explanations

    retreat

    np_max-act · gemini-2.0-flash

    This neuron responds to verbs and phrases describing military defeat or retreat in battle.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    io
    -0.07
    (ind
    -0.07
    [U
    -0.06
    초
    -0.06
    ΙΟ
    -0.06
     ホ
    -0.06
    ission
    -0.06
    _LCD
    -0.05
    (Un
    -0.05
    SetFont
    -0.05
    POSITIVE LOGITS
     pym
    0.07
    Cream
    0.07
     vl
    0.07
     حداقل
    0.07
     prestige
    0.06
    cken
    0.06
    _PARTITION
    0.06
     přesně
    0.06
    .diag
    0.06
     DAN
    0.06
    Act Density 0.024%

    No Known Activations