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

    Neuronpedia

    INDEX
    Explanations

    expressions of personal beliefs or opinions

    oai_token-act-pair · gpt-3.5-turbo

    statements of belief or opinions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    grad
    -0.65
     dry
    -0.60
     spring
    -0.59
     parallel
    -0.59
     port
    -0.58
     stretch
    -0.57
     free
    -0.57
     phase
    -0.57
     length
    -0.56
     idle
    -0.56
    POSITIVE LOGITS
     believes
    3.16
     thinks
    2.10
     believe
    2.04
     understands
    1.92
     expects
    1.91
     believed
    1.82
     considers
    1.75
     insists
    1.73
     disagrees
    1.67
     contends
    1.67
    Act Density 0.014%

    No Known Activations