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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the word "keep" or its variations in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    maintain something

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    PerformLayout
    -0.84
     ***!
    -0.59
    RegressionTest
    -0.59
     BorderRadius
    -0.52
    FieldDescriptor
    -0.51
    photobucket
    -0.49
    erol
    -0.48
     Colla
    -0.48
     Modal
    -0.48
     Walkover
    -0.48
    POSITIVE LOGITS
     keep
    1.32
    keep
    1.32
    Keep
    1.28
     Keep
    1.27
     KEEP
    1.16
    KEEP
    1.16
    Keeping
    1.13
     Keeping
    1.12
     keeping
    1.10
     keeps
    1.07
    Act Density 0.013%

    No Known Activations