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

    Neuronpedia

    INDEX
    Explanations

    instances of time-related phrases indicating the present or recent past

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     bisher
    -0.23
     lately
    -0.21
     recently
    -0.20
     initially
    -0.20
    æľĢåĪĿ
    -0.19
     until
    -0.18
    until
    -0.18
     recent
    -0.18
    kowski
    -0.18
    then
    -0.17
    POSITIVE LOGITS
    adays
    0.24
     still
    0.21
     STILL
    0.20
    aday
    0.20
     Still
    0.19
    Still
    0.18
    still
    0.18
    ä»į
    0.16
     descendant
    0.16
     descendants
    0.15
    Act Density 0.126%

    No Known Activations