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

    Neuronpedia

    INDEX
    Explanations

    keep + ongoing action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    at
    0.82
    ки
    0.70
     adversaries
    0.63
     allies
    0.63
     देखील
    0.63
     coexistence
    0.62
     совмести
    0.62
    ﻋ
    0.62
    जेदार
    0.62
     critters
    0.62
    POSITIVE LOGITS
    Over
    0.75
    紶
    0.71
    嘗試
    0.66
    From
    0.65
    '){
    0.65
    MS
    0.64
    Om
    0.63
    ORDER
    0.63
    ’
    0.63
    END
    0.62
    Act Density 0.012%

    No Known Activations