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

    Neuronpedia

    INDEX
    Explanations

    conversational phrases and expressions that indicate dialogue or thought

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

    Followed by a comma or quote

    np_acts-logits-general · gemini-2.0-flash

    I was like

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    enumi
    -0.79
    )__
    -0.60
     صوتيه
    -0.60
    itespace
    -0.59
    TagHelper
    -0.59
     "/")
    -0.59
    ihnachten
    -0.59
    ★★★★☆
    -0.59
    .。
    -0.57
    ]").
    -0.57
    POSITIVE LOGITS
     Hey
    0.71
    fjspx
    0.70
     давайте
    0.69
     Why
    0.67
     hey
    0.66
     Please
    0.64
     How
    0.63
     Let
    0.62
     please
    0.61
     veuillez
    0.60
    Act Density 0.278%

    No Known Activations