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

    Neuronpedia

    INDEX
    Explanations

    specific time indicators

    oai_token-act-pair · gpt-3.5-turbo

    references to recent events or time-related phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Osc
    -0.70
    holes
    -0.62
    aband
    -0.61
    adle
    -0.61
    plain
    -0.60
     encour
    -0.60
     SOM
    -0.59
     Hers
    -0.59
    obin
    -0.58
    entials
    -0.57
    POSITIVE LOGITS
     month
    1.03
     week
    0.99
     Month
    0.91
     year
    0.89
     summer
    0.79
    ime
    0.79
     Update
    0.79
     ditch
    0.78
     night
    0.76
    fm
    0.76
    Act Density 0.044%

    No Known Activations