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

    Neuronpedia

    INDEX
    Explanations

    introduces new updates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    howto
    0.42
     ווי
    0.41
     Fatty
    0.38
     Mutant
    0.38
     Neuroscience
    0.37
     Wicked
    0.37
    Pete
    0.36
     حوالے
    0.35
    푥
    0.35
    водство
    0.35
    POSITIVE LOGITS
    exit
    0.48
     exit
    0.47
    Exit
    0.45
     exits
    0.41
     Schall
    0.41
     four
    0.40
    rianças
    0.40
     verify
    0.40
     Exit
    0.39
     assumption
    0.39
    Act Density 0.001%

    No Known Activations