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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks, particularly commas, indicating a focus on sentence structure and pauses in thought

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    asm
    -0.17
    which
    -0.16
    combe
    -0.15
    ovsky
    -0.15
    wig
    -0.15
     Serialized
    -0.15
    uggle
    -0.15
    ÌĨ
    -0.14
    rzy
    -0.14
    kiye
    -0.14
    POSITIVE LOGITS
     quite
    0.17
     arg
    0.15
    (Unknown
    0.15
    shall
    0.15
    ünd
    0.15
     Shall
    0.15
     shall
    0.14
     gulp
    0.14
    awi
    0.14
     Klopp
    0.14
    Act Density 0.054%

    No Known Activations