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

    Neuronpedia

    INDEX
    Explanations

    the keyword "End" occurring with a high activation value

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "End," indicating a focus on concluding sections or phrases in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    issance
    -0.84
    kaya
    -0.79
    IGHTS
    -0.68
    PDATE
    -0.67
    RNA
    -0.67
    ppo
    -0.66
    kson
    -0.63
    hee
    -0.63
    chy
    -0.62
    Magikarp
    -0.62
    POSITIVE LOGITS
    angered
    1.22
    angering
    1.21
    owment
    1.09
    urance
    1.05
    ocrin
    1.05
    ocrine
    1.00
    orph
    0.99
    orse
    0.90
    angers
    0.88
    notes
    0.84
    Act Density 0.026%

    No Known Activations