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

    Neuronpedia

    INDEX
    Explanations

    words related to technology and computer networks

    oai_token-act-pair · gpt-3.5-turbo

    instances of circuses or circus-related terminology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Merit
    -0.59
     è£ıç
    -0.58
     Guant
    -0.58
    cigarettes
    -0.54
    QUEST
    -0.52
    LOS
    -0.52
    DCS
    -0.51
    ternity
    -0.50
     constitu
    -0.49
    cffff
    -0.49
    POSITIVE LOGITS
     however
    0.87
     meanwhile
    0.82
     consists
    0.67
     consisted
    0.66
    anwhile
    0.61
     moreover
    0.59
     includes
    0.57
     latter
    0.57
     retains
    0.56
     therefore
    0.55
    Act Density 2.483%

    No Known Activations