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

    Neuronpedia

    INDEX
    Explanations

    phrases related to experimental studies or trials

    oai_token-act-pair · gpt-3.5-turbo

    instances of pilot programs and studies related to innovative projects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    deen
    -0.76
     Secrets
    -0.74
    gew
    -0.73
    clock
    -0.69
    SHARE
    -0.69
    bryce
    -0.68
    mar
    -0.67
    yll
    -0.66
    nick
    -0.66
    gar
    -0.64
    POSITIVE LOGITS
     experiment
    1.17
     experimental
    1.12
     experimented
    1.11
     experiments
    1.09
     experimenting
    1.05
     testing
    0.92
     tested
    0.92
     commercially
    0.89
     experimentation
    0.87
     randomized
    0.87
    Act Density 0.874%

    No Known Activations