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

    Neuronpedia

    INDEX
    Explanations

    past tense forms of "have" and related phrases indicating possession or existence

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

    always have/be/had

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     agricole
    -0.40
    médiaire
    -0.35
    terung
    -0.35
     vuestros
    -0.35
     yapılan
    -0.35
     vuestro
    -0.35
    NSURL
    -0.34
     varios
    -0.33
     Wör
    -0.33
     brief
    -0.33
    POSITIVE LOGITS
    Always
    1.02
     Always
    0.97
    always
    0.95
    ALWAYS
    0.93
     ALWAYS
    0.92
     always
    0.88
     alway
    0.80
    siempre
    0.74
     zawsze
    0.73
     Siempre
    0.72
    Act Density 0.015%

    No Known Activations