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

    Neuronpedia

    INDEX
    Explanations

    phrases that repeat or describe recurring actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to repetition or reiteration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    minster
    -0.74
    Brave
    -0.73
    ardless
    -0.71
    behind
    -0.67
    vana
    -0.66
    uala
    -0.65
     Squadron
    -0.62
    Bu
    -0.61
    nas
    -0.61
    emouth
    -0.59
    POSITIVE LOGITS
    hang
    0.87
     hump
    0.85
    ealous
    0.83
    whelming
    0.81
     fence
    0.74
    arching
    0.72
     weekend
    0.71
    whel
    0.70
    drive
    0.70
     hundred
    0.68
    Act Density 0.099%

    No Known Activations