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

    Neuronpedia

    INDEX
    Explanations

    code and math

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     loophole
    -0.07
     zp
    -0.06
     вб
    -0.06
    ็นท
    -0.06
    .Does
    -0.06
    .document
    -0.06
     لي
    -0.06
     лим
    -0.06
    "P
    -0.06
    。その
    -0.06
    POSITIVE LOGITS
    Signature
    0.07
    aal
    0.07
     fname
    0.07
    	mem
    0.06
    Never
    0.06
    -dismiss
    0.06
     Basis
    0.06
    си
    0.06
     Caller
    0.06
     Tap
    0.06
    Act Density 0.003%

    No Known Activations