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

    Neuronpedia

    INDEX
    Explanations

    repeated phrases that indicate relationships or actions related to "doing" something

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

    explaining reasons or associations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     pérd
    -0.61
     jäsen
    -0.52
    Dış
    -0.52
     näin
    -0.50
     econó
    -0.49
    ientras
    -0.48
     asisti
    -0.48
     patrulla
    -0.48
     clín
    -0.47
     hänen
    -0.47
    POSITIVE LOGITS
    ToDo
    0.54
    ioutil
    0.52
     relate
    0.52
     about
    0.50
     related
    0.50
     związane
    0.49
    ImageContext
    0.49
     gin
    0.48
     Gin
    0.48
     relating
    0.48
    Act Density 0.007%

    No Known Activations