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

    Neuronpedia

    INDEX
    Explanations

    words related to intensity or emphasis

    oai_token-act-pair · gpt-3.5-turbo

    sequences of letters with high frequency or certain phonetic patterns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     domains
    -0.83
     assemblies
    -0.81
     publications
    -0.81
     fixtures
    -0.81
     establishments
    -0.78
     exhibits
    -0.77
     complexes
    -0.77
     exploits
    -0.76
     clients
    -0.76
    Ĥİ
    -0.76
    POSITIVE LOGITS
    ighter
    0.98
    ounding
    0.96
    iddling
    0.95
    icer
    0.90
    attering
    0.90
    ittance
    0.89
    atical
    0.89
    abyte
    0.89
    auc
    0.88
    aser
    0.87
    Act Density 0.230%

    No Known Activations