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

    Neuronpedia

    INDEX
    Explanations

    sustains or keeping alive

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Invite
    0.39
    呈
    0.38
    ذف
    0.38
     invitación
    0.38
    进一步
    0.37
     opening
    0.36
    的影响
    0.36
     सीख
    0.36
     invite
    0.35
     Opening
    0.35
    POSITIVE LOGITS
     sustain
    1.80
     sustaining
    1.80
     sustains
    1.80
     sustained
    1.51
    维持
    1.50
     Sust
    1.48
    sustaining
    1.45
     sust
    1.40
     keeping
    1.39
     sosten
    1.35
    Act Density 0.027%

    No Known Activations