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

    Neuronpedia

    INDEX
    Explanations

    instances where the word "probably" is used

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "probably" indicating uncertainty or speculation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    issy
    -0.83
    gencies
    -0.78
    lette
    -0.74
    eller
    -0.72
    elight
    -0.72
    lain
    -0.71
    uality
    -0.71
    yer
    -0.71
    leground
    -0.70
    ritch
    -0.70
    POSITIVE LOGITS
     incent
    0.82
    Ń·
    0.75
    ©¶æ
    0.74
     ali
    0.74
     infer
    0.74
     assume
    0.73
     gonna
    0.72
     preferring
    0.71
     misunder
    0.70
     occas
    0.67
    Act Density 0.017%

    No Known Activations