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

    Neuronpedia

    INDEX
    Explanations

    conditional phrases that express exceptions or nuances

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

    "that doesn't mean"

    np_acts-logits-general · gemini-2.0-flash

    that doesn't mean

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     otherwise
    -0.45
     un
    -0.44
    Un
    -0.42
    unwrap
    -0.42
    否则
    -0.41
    un
    -0.41
    Detailed
    -0.40
    LogFactory
    -0.39
    Pt
    -0.39
     topped
    -0.38
    POSITIVE LOGITS
    tagHelperRunner
    0.93
     kasarigan
    0.90
     Roskov
    0.84
     heißt
    0.79
     znaczy
    0.78
    észetes
    0.76
    withstanding
    0.76
     للاسماء
    0.75
     betyr
    0.73
     ivelany
    0.73
    Act Density 0.136%

    No Known Activations