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

    Neuronpedia

    INDEX
    Explanations

    what i and what if

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    (,
    0.41
     温
    0.37
     هد
    0.37
    Yet
    0.36
     Yet
    0.36
    ,),
    0.36
    Extended
    0.36
    *',
    0.36
    руют
    0.35
     آد
    0.35
    POSITIVE LOGITS
     does
    0.56
     sets
    0.50
     distinguishes
    0.50
     do
    0.50
     exactly
    0.49
     defines
    0.48
     separates
    0.48
     complicates
    0.47
    does
    0.47
    kaç
    0.46
    Act Density 0.035%

    No Known Activations