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

    Neuronpedia

    INDEX
    Explanations

    transitional phrases and qualifiers used to indicate relationships or conditions in arguments

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

    preceding certain conjunctions

    np_acts-logits-general · gemini-2.0-flash

    conclusive connective words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     vectorielle
    -0.54
     feroit
    -0.54
     publiques
    -0.52
     lourd
    -0.50
     valiente
    -0.50
     conmigo
    -0.49
     cabra
    -0.47
     constitucional
    -0.47
     coû
    -0.47
     comigo
    -0.47
    POSITIVE LOGITS
    tagHelperRunner
    0.52
    期刊论文
    0.52
    وفاته
    0.49
     elif
    0.48
    }{*}{
    0.48
     surla
    0.48
    omiast
    0.48
     betweenstory
    0.48
    وصلات
    0.47
     snippetHide
    0.46
    Act Density 0.290%

    No Known Activations