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

    Neuronpedia

    INDEX
    Explanations

    sentences, especially those that come towards the end of a passage or include transitions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "sentence."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    BLIC
    -0.85
    ikarp
    -0.85
    rists
    -0.79
    jri
    -0.79
    eus
    -0.75
    rament
    -0.75
    IED
    -0.75
    roxy
    -0.74
    enhagen
    -0.74
    owered
    -0.73
    POSITIVE LOGITS
     uttered
    1.18
     sentences
    1.02
     phrases
    0.95
     paragraphs
    0.95
     sentence
    0.92
     quotation
    0.89
    mith
    0.84
     paragraph
    0.83
     text
    0.83
     comprehension
    0.83
    Act Density 0.054%

    No Known Activations