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

    Neuronpedia

    INDEX
    Explanations

    words related to order, organization, or aesthetics

    oai_token-act-pair · gpt-3.5-turbo

    expressions that convey pleasantness or satisfaction in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Keys
    -0.66
    gate
    -0.65
    orno
    -0.63
    illary
    -0.62
     Victims
    -0.61
     Citizens
    -0.61
     necessity
    -0.60
    anish
    -0.60
     verbs
    -0.60
     Responsibility
    -0.60
    POSITIVE LOGITS
    baum
    0.92
     suited
    0.86
     exting
    0.75
     illustrates
    0.74
    ãĤ©
    0.73
    ufact
    0.73
     illustrated
    0.72
    ength
    0.71
     BELOW
    0.71
     adapted
    0.70
    Act Density 0.013%

    No Known Activations