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

    Neuronpedia

    INDEX
    Explanations

    sequences of numbers and symbols that resemble special characters like arrows and lines

    oai_token-act-pair · gpt-3.5-turbo

    sequences of special characters and their patterns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     behav
    -0.72
     Canary
    -0.72
     Afric
    -0.71
     fed
    -0.64
     Leilan
    -0.63
     association
    -0.63
    apult
    -0.61
     dressing
    -0.61
     protector
    -0.58
     lax
    -0.58
    POSITIVE LOGITS
    percent
    1.14
    renheit
    0.96
    undred
    0.96
     inches
    0.86
     increments
    0.86
    lbs
    0.86
    degree
    0.86
     ounces
    0.85
    pound
    0.84
    iversary
    0.83
    Act Density 0.083%

    No Known Activations