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

    Neuronpedia

    INDEX
    Explanations

    news-related discussions and reports from various sources

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     ç¥ŀ
    -0.85
    avorite
    -0.85
    animate
    -0.83
    lain
    -0.80
    icidal
    -0.77
    artney
    -0.75
    luster
    -0.75
     Valhalla
    -0.74
    wagen
    -0.74
    activation
    -0.74
    POSITIVE LOGITS
     columnist
    1.35
     Reporter
    1.20
     News
    1.19
     reporter
    1.19
     correspondent
    1.17
     reporters
    1.15
    News
    1.15
     Inqu
    1.13
     editorial
    1.12
     newspaper
    1.11
    Act Density 3.452%

    No Known Activations