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

    Neuronpedia

    INDEX
    Explanations

    references to conditional phrases or clauses

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

    "As" followed by a subject or clause

    np_acts-logits-general · gemini-2.0-flash

    As introductions to clauses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     AssemblyCulture
    -0.68
     estando
    -0.63
    apimachinery
    -0.63
     habiendo
    -0.62
     HttpNotFound
    -0.60
    astify
    -0.58
    gameserver
    -0.56
     Мексичка
    -0.56
     vocale
    -0.55
     suyo
    -0.53
    POSITIVE LOGITS
     they
    0.99
     we
    0.91
     temperatures
    0.74
     soon
    0.72
     hundreds
    0.71
     thousands
    0.70
    脚注の使い方
    0.69
     each
    0.69
     he
    0.66
     she
    0.66
    Act Density 0.098%

    No Known Activations