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

    Neuronpedia

    INDEX
    Explanations

    variable assignments and labels

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    @@]
    0.38
    лася
    0.37
     برق
    0.36
    也会
    0.35
    iftoire
    0.35
    ⤵
    0.34
    νοντας
    0.34
     balsam
    0.33
    otherArchive
    0.33
    σταν
    0.33
    POSITIVE LOGITS
     The
    0.51
     represents
    0.49
     =
    0.47
    The
    0.47
     Represents
    0.42
        
    0.42
     ise
    0.42
    0.39
     or
    0.39
     denotes
    0.39
    Act Density 0.101%

    No Known Activations