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

    Neuronpedia

    INDEX
    Explanations

    model respond with phrase

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fumes
    0.48
     அறை
    0.46
    stig
    0.45
    filtering
    0.45
    often
    0.44
    soff
    0.44
    text
    0.43
    ujesz
    0.43
    impact
    0.43
    quorum
    0.43
    POSITIVE LOGITS
     Pt
    0.43
    нию
    0.42
    inguish
    0.40
     Endpoint
    0.40
     parallax
    0.40
     MUI
    0.40
     Ukraine
    0.39
    жет
    0.39
     HOW
    0.39
     nasz
    0.39
    Act Density 0.001%

    No Known Activations