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

    Neuronpedia

    INDEX
    Explanations

    characters in a non-Latin script known as Gothic letters

    oai_token-act-pair · gpt-3.5-turbo

    special characters or unusual symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    notes
    -0.67
    espie
    -0.65
     reminder
    -0.64
     Tribune
    -0.64
    ITNESS
    -0.63
     stake
    -0.63
     plain
    -0.62
     Plain
    -0.60
     Deep
    -0.60
     memorandum
    -0.59
    POSITIVE LOGITS
    ĥ
    1.45
    ½
    1.07
    é¾į
    1.04
    ÏĦ
    1.02
    ¼
    1.01
    Ĥ
    0.99
    Ī
    0.96
    ħ
    0.95
    Ùħ
    0.93
    º
    0.93
    Act Density 0.009%

    No Known Activations