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

    Neuronpedia

    INDEX
    Explanations

    question

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    τοκ
    -0.07
    nth
    -0.06
     �
    -0.06
    �
    -0.06
     Wolverine
    -0.06
    In
    -0.06
    dale
    -0.06
    (cors
    -0.06
     ç
    -0.06
    "I
    -0.06
    POSITIVE LOGITS
    thought
    0.07
     referring
    0.07
     Figure
    0.06
     curiosity
    0.06
    Next
    0.06
    =target
    0.06
     question
    0.06
     selecting
    0.06
    trained
    0.06
     Ogre
    0.06
    Act Density 0.141%

    No Known Activations