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

    Neuronpedia

    INDEX
    Explanations

    phrases and conjunctions that emphasize connections or repetitive concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     timelines
    -0.17
     timed
    -0.16
    igon
    -0.16
     timely
    -0.15
    lee
    -0.15
     timings
    -0.14
     tarde
    -0.14
    eyh
    -0.14
     Timeline
    -0.14
     surv
    -0.14
    POSITIVE LOGITS
    åľ°çĤ¹
    0.19
     effort
    0.17
     location
    0.17
    é¢ij
    0.17
     dime
    0.17
     expense
    0.16
    åĪ»
    0.16
     frequency
    0.15
    åł´æīĢ
    0.15
    oda
    0.15
    Act Density 0.086%

    No Known Activations