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

    Neuronpedia

    INDEX
    Explanations

    Android code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     MET
    -0.06
    (alert
    -0.06
    فته
    -0.06
     allowances
    -0.06
    .leave
    -0.06
    шили
    -0.06
     Mex
    -0.06
     이야
    -0.06
    uencia
    -0.06
     Manor
    -0.05
    POSITIVE LOGITS
    ,
↵
    0.07
    очные
    0.07
     разв
    0.07
     saf
    0.07
    méně
    0.07
     pants
    0.07
     comics
    0.07
    .jpeg
    0.06
     multi
    0.06
    cer
    0.06
    Act Density 0.004%

    No Known Activations