© 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. 22-GEMMASCOPE-RES-131K
    7. 74016
    Prev
    Next
    INDEX
    Explanations

    months and years

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

    The neuron detects tokens that form temporal expressions—especially months, days, years, and time stamps.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    Zubereitung
    -1.01
    Earlier
    -0.98
     préc
    -0.94
     kän
    -0.93
    roix
    -0.93
     berlebihan
    -0.92
    Dati
    -0.92
     Dí
    -0.91
     komme
    -0.90
     Gelen
    -0.90
    POSITIVE LOGITS
    zulegen
    0.94
     Türen
    0.92
    and
    0.91
    it
    0.91
     derselben
    0.90
     отредактировал
    0.88
     Umständen
    0.87
     demselben
    0.86
    pieza
    0.85
     it
    0.85
    Activations Density 0.074%

    No Known Activations