© 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. 98397
    Prev
    Next
    INDEX
    Explanations

    time references and past events

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

    The neuron primarily fires on numeric tokens (years, dates, counts, measurements, percentages, etc.).

    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
     продолжает
    -1.03
     continue
    -0.91
     weiterhin
    -0.89
     nuevas
    -0.88
     stále
    -0.88
     شدن
    -0.87
     continues
    -0.85
     bijzondere
    -0.84
     Continuing
    -0.83
     coraz
    -0.82
    POSITIVE LOGITS
     when
    1.08
     comeback
    1.00
     Utilizamos
    0.95
     déjà
    0.91
     brosse
    0.90
    间隔
    0.90
     but
    0.90
     effekt
    0.88
    sebagai
    0.88
     years
    0.88
    Activations Density 0.053%

    No Known Activations