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

    Neuronpedia

    INDEX
    Explanations

    capitalized words with a mixture of special characters, potentially representing an encoding or foreign language

    oai_token-act-pair · gpt-3.5-turbo

    elements related to obscure or non-standard characters and symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     juggling
    -0.79
     Wilmington
    -0.79
     listing
    -0.77
     bouncing
    -0.74
     loopholes
    -0.71
     sorting
    -0.71
     boosting
    -0.70
     Collider
    -0.68
     superheroes
    -0.68
     momentum
    -0.68
    POSITIVE LOGITS
    ï¸ı
    1.33
    uv
    1.21
    и
    1.20
    estern
    1.19
    ansk
    1.19
    ti
    1.19
    ô
    1.18
    mia
    1.17
    nai
    1.17
    nia
    1.16
    Act Density 0.089%

    No Known Activations