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

    Neuronpedia

    INDEX
    Explanations

    these

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     liar
    -0.07
     }],↵
    -0.07
    /models
    -0.07
    .FormStartPosition
    -0.06
     Tokyo
    -0.06
    	Title
    -0.06
     Soy
    -0.06
     beach
    -0.06
     Brotherhood
    -0.06
    	first
    -0.06
    POSITIVE LOGITS
    tems
    0.07
    Formatted
    0.06
    KNOWN
    0.06
     помощ
    0.06
     pensions
    0.06
    INATION
    0.06
    rollback
    0.06
     Overwatch
    0.06
    esson
    0.06
    فصل
    0.06
    Act Density 0.066%

    No Known Activations