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

    Neuronpedia

    INDEX
    Explanations

    suggesting actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    €™
    0.52
     veggie
    0.49
     și
    0.49
     magical
    0.49
     iter
    0.48
    -
    0.47
     và
    0.46
     trial
    0.46
     pues
    0.46
    л
    0.45
    POSITIVE LOGITS
    ۔
    0.53
     આપી
    0.48
    这部
    0.48
    giving
    0.48
    ündung
    0.47
     devoting
    0.47
     exchanging
    0.47
    removing
    0.47
     دادن
    0.46
     aumentando
    0.46
    Act Density 0.008%

    No Known Activations