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

    Neuronpedia

    INDEX
    Explanations

    say "the"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     whe
    -0.07
     lí
    -0.06
    ān
    -0.06
     takes
    -0.06
     unb
    -0.06
    	dialog
    -0.06
    arehouse
    -0.06
     тран
    -0.06
     painfully
    -0.06
    _ct
    -0.06
    POSITIVE LOGITS
    顶
    0.07
     contrib
    0.06
     καθώς
    0.06
    /N
    0.06
     trivia
    0.06
    acic
    0.06
     GmbH
    0.06
     todos
    0.06
    _npc
    0.06
     Flu
    0.06
    Act Density 0.016%

    No Known Activations