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

    Neuronpedia

    INDEX
    Explanations

    executing data tasks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     wasn
    0.41
     sarebbe
    0.40
     weren
    0.38
     Haziran
    0.38
     était
    0.38
     tuvieron
    0.38
     puisqu
    0.37
    ﻝ
    0.37
     Throughout
    0.36
     essendo
    0.36
    POSITIVE LOGITS
     the
    0.61
     various
    0.51
     my
    0.50
     histograms
    0.50
    各种
    0.48
     JSON
    0.47
     an
    0.46
    /
    0.46
     certain
    0.46
     различные
    0.46
    Act Density 0.043%

    No Known Activations