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

    Neuronpedia

    INDEX
    Explanations

    occurrences where a statement is completed or emphasized

    oai_token-act-pair · gpt-3.5-turbo

    periods at the end of sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     satell
    -0.95
     carbohyd
    -0.87
     challeng
    -0.87
     millenn
    -0.84
    ¥ŀ
    -0.82
     charact
    -0.81
     strugg
    -0.81
     neighb
    -0.78
     transitions
    -0.78
     tremend
    -0.78
    POSITIVE LOGITS
     Presumably
    1.65
     Apparently
    1.64
     Basically
    1.52
     Which
    1.51
     Seems
    1.47
     Nope
    1.46
     Hmm
    1.42
     Sounds
    1.42
     Yep
    1.41
     Turns
    1.41
    Act Density 0.565%

    No Known Activations