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

    Neuronpedia

    INDEX
    Explanations

    the word "that" at various activations

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "that" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    erenn
    -0.58
    ogether
    -0.57
    Ire
    -0.56
    aukee
    -0.52
    legate
    -0.52
    Guard
    -0.52
    izont
    -0.51
    raq
    -0.51
    idan
    -0.50
     Coordinator
    -0.50
    POSITIVE LOGITS
     fateful
    0.55
     contradicts
    0.55
    esson
    0.54
     Xiaomi
    0.52
     pesky
    0.52
     accompanies
    0.52
    advertisement
    0.51
    lav
    0.51
     IMAGES
    0.50
    ihad
    0.50
    Act Density 0.268%

    No Known Activations