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

    Neuronpedia

    INDEX
    Explanations

    common utilities and libraries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     appropriate
    0.93
     suitable
    0.86
     descendant
    0.82
     OK
    0.82
     making
    0.81
     heir
    0.80
     limited
    0.80
     AND
    0.80
     like
    0.80
     appropriately
    0.80
    POSITIVE LOGITS
    util
    1.16
    utils
    1.12
    common
    1.08
    api
    1.00
    exceptions
    0.98
    utilities
    0.95
    examples
    0.95
    runtime
    0.94
    exception
    0.93
    sdk
    0.93
    Act Density 0.004%

    No Known Activations