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

    Neuronpedia

    INDEX
    Explanations

    words that are prepositions, conjunctions, some forms of "to be," or that follow prepositions, suggesting that it finds grammatical connections between words.

    oai_token-act-pair · gemini-2.0-flash

    terms related to respect and regard

    np_token-act-pair-logits · gpt-4o-mini

    with

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Pautan
    -0.58
     cuales
    -0.50
     voulu
    -0.48
     kdy
    -0.46
     środ
    -0.46
    PreferredItem
    -0.45
    gdx
    -0.45
     Whether
    -0.44
     Stunde
    -0.43
    <u>
    -0.43
    POSITIVE LOGITS
     respect
    1.11
     impunity
    1.06
     regard
    1.05
     regards
    1.02
    regard
    0.91
     ease
    0.91
    respect
    0.87
    inthe
    0.84
    ]--;
    0.82
    outa
    0.81
    Act Density 16.235%

    No Known Activations