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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating quantity or plurality

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating quantity or frequency of things

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anism
    -0.77
    arta
    -0.76
     srfAttach
    -0.76
    assisted
    -0.76
    ENCE
    -0.75
    slaught
    -0.73
    matter
    -0.73
    buster
    -0.71
    ãĥ¼ãĥ«
    -0.70
    escription
    -0.69
    POSITIVE LOGITS
     exceptions
    1.23
     reasons
    1.22
     indications
    1.17
     occasions
    1.12
     ways
    1.11
     differences
    1.08
     instances
    1.07
     similarities
    1.05
     caveats
    0.95
     discrepancies
    0.95
    Act Density 0.165%

    No Known Activations