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

    Neuronpedia

    INDEX
    Explanations

    time-related phrases mentioning specific durations such as "five years" or "seven years"

    oai_token-act-pair · gpt-3.5-turbo

    references to time spans or durations related to years and days

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inyl
    -0.73
    ilde
    -0.66
    arag
    -0.66
    abeth
    -0.66
    reditary
    -0.65
    atech
    -0.65
    romeda
    -0.65
    asper
    -0.64
    abilia
    -0.64
     ali
    -0.63
    POSITIVE LOGITS
     span
    0.92
     increments
    0.86
     ago
    0.86
     stride
    0.81
    thood
    0.79
     nutshell
    0.78
    åŃIJ
    0.75
     thereafter
    0.75
    ²¾
    0.70
    Shape
    0.70
    Act Density 0.133%

    No Known Activations