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

    Neuronpedia

    INDEX
    Explanations

    speaker names in dialogue

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    System
    0.34
    вача
    0.32
    robust
    0.31
    system
    0.30
    defaults
    0.30
    valid
    0.30
    prime
    0.29
    及其
    0.29
    Users
    0.29
     இவரது
    0.29
    POSITIVE LOGITS
     Nick
    0.35
     onlookers
    0.32
    兩人
    0.32
     Cheryl
    0.30
     Christian
    0.30
     Bowling
    0.29
     Harry
    0.29
     Beach
    0.29
     Jean
    0.28
     Jon
    0.28
    Act Density 0.013%

    No Known Activations