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

    Neuronpedia

    INDEX
    Explanations

    instances of the exclamation "Ha" indicating laughter or surprise, often in various stylings or repetitions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lec
    -0.16
    pong
    -0.15
    cross
    -0.15
    stroy
    -0.15
    212
    -0.14
     Wid
    -0.14
    h
    -0.14
    jets
    -0.14
    horn
    -0.14
    wid
    -0.14
    POSITIVE LOGITS
    unted
    0.24
     ha
    0.24
    iku
    0.23
     Ha
    0.22
    Ha
    0.21
    ifax
    0.21
    ifa
    0.20
    unting
    0.20
    iley
    0.19
    emat
    0.18
    Act Density 0.011%

    No Known Activations