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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 23793
    Prev
    Next
    INDEX
    Explanations

    durations of time

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

    The neuron primarily fires on numeric tokens that denote time—especially years or duration measurements.

    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
     Gläub
    -0.90
    immediately
    -0.88
    鹀
    -0.88
    Chaque
    -0.84
     instantaneously
    -0.84
     dimaksud
    -0.83
     événement
    -0.82
    까지
    -0.80
    まで
    -0.80
     каждый
    -0.79
    POSITIVE LOGITS
     time
    2.13
     timeframe
    1.27
    flat
    1.23
     flat
    1.23
     tijd
    1.22
     from
    1.22
    time
    1.19
     span
    1.16
    FLAT
    1.11
     after
    1.09
    Activations Density 0.038%

    No Known Activations