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

    Neuronpedia

    INDEX
    Explanations

    years followed by dots

    oai_token-act-pair · gpt-3.5-turbo

    instances of commas and related punctuation marks in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    =#
    -0.73
    oll
    -0.71
    "]=>
    -0.71
    ppo
    -0.70
    urry
    -0.69
    ĸ
    -0.69
    ":-
    -0.68
    Definition
    -0.68
    orem
    -0.66
    chin
    -0.66
    POSITIVE LOGITS
     earning
    1.14
     specializing
    1.10
     preferring
    1.08
     oversaw
    1.03
     befriend
    1.00
     including
    0.99
     culminating
    0.99
     overseeing
    0.98
     graduating
    0.97
     assisting
    0.96
    Act Density 0.320%

    No Known Activations