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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Handler
    -0.07
    _root
    -0.07
     discrim
    -0.07
     invites
    -0.06
     अग
    -0.06
    Lt
    -0.06
     fox
    -0.06
     finger
    -0.06
     bras
    -0.06
     cst
    -0.06
    POSITIVE LOGITS
    ↵
↵
↵
↵
    0.07
    iện
    0.07
        
↵
↵
    0.07
     هنر
    0.07
    irq
    0.06
     заяв
    0.06
    ुबह
    0.06
    exels
    0.06
     kişilerin
    0.06
    PathComponent
    0.06
    Act Density 0.004%

    No Known Activations