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

    Neuronpedia

    INDEX
    Explanations

    instances of greeting phrases or casual conversational openings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    AfterClass
    -0.72
     šķ
    -0.61
    hicle
    -0.59
     Barbosa
    -0.57
    eraard
    -0.57
     Faber
    -0.56
    oficina
    -0.56
     Brunner
    -0.56
    wic
    -0.55
     schul
    -0.54
    POSITIVE LOGITS
     Hey
    2.40
    Hey
    2.29
     hey
    2.25
     HEY
    2.23
    hey
    2.02
    HEY
    1.94
     Heywood
    1.38
    Heya
    1.19
    嘿
    1.13
    heyd
    1.10
    Act Density 0.064%

    No Known Activations