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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    reason
    -0.07
    .Do
    -0.06
     persec
    -0.06
    (duration
    -0.06
     extravag
    -0.06
    (fh
    -0.06
    ố
    -0.06
    δει
    -0.06
     ###
    -0.06
     spots
    -0.06
    POSITIVE LOGITS
    之后
    0.07
     jewelry
    0.07
     �
    0.06
    .Interfaces
    0.06
     competit
    0.06
    差
    0.06
    使用
    0.06
     semiclass
    0.06
    uede
    0.06
    نی
    0.06
    Act Density 0.011%

    No Known Activations