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

    Neuronpedia

    INDEX
    Explanations

    document metadata

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     veggies
    0.52
     explanation
    0.50
     celebs
    0.50
     critters
    0.47
     probs
    0.46
     segue
    0.44
     nerd
    0.44
     techs
    0.44
     explicación
    0.43
     parse
    0.42
    POSITIVE LOGITS
    ISBN
    0.81
     Revision
    0.78
     ISBN
    0.77
    Copyright
    0.75
    Disclaimer
    0.73
    Revision
    0.73
     Disclaimer
    0.70
     Copyright
    0.70
     Изда
    0.70
    CONTENTS
    0.69
    Act Density 0.007%

    No Known Activations