© 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. 75353
    Prev
    Next
    INDEX
    Explanations

    months and dates

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

    The neuron primarily activates on numeric tokens (digits or numbers, e.g. dates, counts, or other numerals).

    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
    結果は
    -1.05
    ýš
    -1.04
     spus
    -0.99
    Тел
    -0.99
    nehå
    -0.95
     sahiptir
    -0.94
     successor
    -0.94
     susp
    -0.93
     первая
    -0.93
    <bos>
    -0.92
    POSITIVE LOGITS
     lendemain
    1.04
     könnten
    1.04
     September
    1.02
     August
    0.98
    Aiheesta
    0.93
     October
    0.93
     December
    0.90
     italienischen
    0.89
    
    0.87
     April
    0.87
    Activations Density 0.023%

    No Known Activations