© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 77699
    Prev
    Next
    INDEX
    Explanations

    nighttime activities or periods

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on wording about spending or staying the night (i.e. references to “spend the night,” “stay the night,” and similar overnight contexts).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     today
    -1.08
     oggi
    -0.98
    Day
    -0.93
    today
    -0.91
    Today
    -0.91
     Noon
    -0.88
     Today
    -0.87
     σήμερα
    -0.87
     midday
    -0.85
    day
    -0.84
    POSITIVE LOGITS
     night
    4.50
     nights
    3.34
    Night
    2.78
     Night
    2.77
    night
    2.73
     noite
    2.44
     noche
    2.44
     NIGHT
    2.42
     ночь
    2.38
    夜
    2.38
    Activations Density 0.046%

    No Known Activations