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

    Neuronpedia

    INDEX
    Explanations

    prepositional phrases beginning with 'while'

    oai_token-act-pair · gpt-3.5-turbo

    the word "while" used in contrasting or comparative contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hack
    -0.82
    aer
    -0.81
    iotic
    -0.79
    iot
    -0.76
    enter
    -0.74
    entry
    -0.73
    elle
    -0.72
    atter
    -0.72
    illet
    -0.71
    red
    -0.70
    POSITIVE LOGITS
     retaining
    0.95
     maintaining
    0.93
     ignoring
    0.88
     browsing
    0.88
     simultaneously
    0.84
     preserving
    0.82
     awaiting
    0.80
     acknowledging
    0.79
     keeping
    0.78
     touring
    0.78
    Act Density 0.049%

    No Known Activations