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

    Neuronpedia

    INDEX
    Explanations

    phrases related to numeric values being below or above certain thresholds

    oai_token-act-pair · gpt-3.5-turbo

    references to temperature thresholds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    BUG
    -0.69
    VPN
    -0.67
    Commerce
    -0.65
    NetMessage
    -0.65
    mail
    -0.63
    âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦
    -0.61
     VIDE
    -0.60
    UID
    -0.60
    ãĤ¹ãĥĪ
    -0.60
    oga
    -0.59
    POSITIVE LOGITS
     expectations
    1.09
     normal
    1.02
     expectation
    0.95
    normal
    0.93
     parity
    0.91
     optimum
    0.88
     subsistence
    0.88
     baseline
    0.86
     norms
    0.84
     usual
    0.84
    Act Density 0.124%

    No Known Activations