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

    Neuronpedia

    INDEX
    Explanations

    Code/script excerpts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    upiter
    -0.08
     implicitly
    -0.08
    ista
    -0.07
     sheltered
    -0.07
     shelter
    -0.07
     JWT
    -0.07
     metam
    -0.07
     centre
    -0.07
     separate
    -0.07
     disreg
    -0.07
    POSITIVE LOGITS
     Dodge
    0.08
     Ellis
    0.08
     huit
    0.08
     Examiner
    0.08
    包括
    0.08
     Pour
    0.08
     cancelar
    0.08
     κιν
    0.08
     anciens
    0.07
    +xml
    0.07
    Act Density 0.008%

    No Known Activations