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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing specific measurements or numerical data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ekl
    -0.16
    áng
    -0.16
    ondon
    -0.15
    лл
    -0.15
    ioni
    -0.14
    òn
    -0.13
    .sky
    -0.13
    avig
    -0.13
    abar
    -0.13
    cken
    -0.13
    POSITIVE LOGITS
     specifically
    0.81
     Specifically
    0.77
    specific
    0.65
    Specific
    0.61
     especÃŃf
    0.54
     specific
    0.51
    åħ·ä½ĵ
    0.49
    pecific
    0.47
    -specific
    0.45
     конкÑĢеÑĤ
    0.43
    Act Density 0.221%

    No Known Activations