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

    Neuronpedia

    INDEX
    Explanations

    words related to cycles or recurring patterns

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to cycles or repetitive processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inez
    -0.93
    essors
    -0.87
    imus
    -0.84
    amina
    -0.84
    avez
    -0.84
    orage
    -0.79
    roit
    -0.79
    awaru
    -0.78
    initions
    -0.78
    ractive
    -0.76
    POSITIVE LOGITS
     cycle
    1.14
     Cycle
    0.95
     cycles
    0.94
    cycle
    0.91
    cycles
    0.90
    cles
    0.84
     erg
    0.83
     repeats
    0.80
     alternating
    0.75
     Delay
    0.73
    Act Density 0.022%

    No Known Activations