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

    Neuronpedia

    INDEX
    Explanations

    configuration parameters and settings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Estas
    0.48
    Energie
    0.46
    SError
    0.44
    Estado
    0.43
    عبد
    0.42
     העולם
    0.42
    גי
    0.41
     morrer
    0.41
    Estos
    0.41
    Editar
    0.41
    POSITIVE LOGITS
     activating
    0.50
     thresholds
    0.47
     parameters
    0.47
     activate
    0.45
     settings
    0.45
     Specify
    0.45
     operated
    0.44
     with
    0.43
     upstream
    0.43
     Settings
    0.42
    Act Density 0.054%

    No Known Activations