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

    Neuronpedia

    INDEX
    Explanations

    .

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    יים
    0.40
    able
    0.39
    oit
    0.38
    oqu
    0.38
    éments
    0.38
    em
    0.38
     Technical
    0.37
     -
    0.36
    점
    0.36
    Télé
    0.35
    POSITIVE LOGITS
     universitario
    0.53
     estaban
    0.49
     puso
    0.48
     estavam
    0.46
     cuja
    0.46
     hice
    0.45
     liel
    0.45
     aveva
    0.45
     aproximadamente
    0.45
    들에게
    0.45
    Act Density 0.008%

    No Known Activations