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

    Neuronpedia

    INDEX
    Explanations

    percentage increases or decreases

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating measurements or statistics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Emails
    -0.73
     airs
    -0.69
     Said
    -0.66
    NetMessage
    -0.65
    letters
    -0.64
    eston
    -0.63
    ocom
    -0.62
    items
    -0.62
    udic
    -0.61
    went
    -0.61
    POSITIVE LOGITS
     ratio
    1.12
     fraction
    1.11
     figure
    1.05
     feat
    1.04
     underestimate
    0.99
     statistic
    0.98
     decrease
    0.97
     percentage
    0.96
     margin
    0.96
     threshold
    0.95
    Act Density 0.123%

    No Known Activations