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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Sentences after periods

    np_acts-logits-general · gemini-2.0-flash

    causal or logical connectors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     And
    -0.97
    And
    -0.94
     You
    -0.93
    You
    -0.88
     you
    -0.87
     YOU
    -0.79
     Your
    -0.78
    Your
    -0.77
     Yeah
    -0.77
     Maybe
    -0.76
    POSITIVE LOGITS
    由於
    0.78
    由于
    0.77
     Appropriate
    0.77
     Additional
    0.75
     Care
    0.75
    Appropriate
    0.74
     Careful
    0.73
     Efforts
    0.72
     Considerable
    0.72
     caution
    0.70
    Act Density 0.939%

    No Known Activations