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

    Neuronpedia

    INDEX
    Explanations

    bullet points or lists

    oai_token-act-pair · gpt-3.5-turbo

    sentences that contain full stops (periods)

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     affili
    -0.81
     fraternity
    -0.81
     purse
    -0.76
     unheard
    -0.76
     footing
    -0.75
     enthusi
    -0.74
     ethic
    -0.74
     apparel
    -0.73
     friendships
    -0.73
     cradle
    -0.73
    POSITIVE LOGITS
     Additionally
    1.35
     However
    1.35
     Therefore
    1.33
     Alternatively
    1.32
     Otherwise
    1.27
     Unfortunately
    1.26
     Currently
    1.26
     Moreover
    1.24
     Specifically
    1.24
     Furthermore
    1.23
    Act Density 0.379%

    No Known Activations