© 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. 22-GEMMASCOPE-RES-131K
    7. 64689
    Prev
    Next
    INDEX
    Explanations

    dates and numbers

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

    The neuron strongly activates on numeric tokens—particularly four-digit year references (dates).

    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
    Géographie
    -1.05
    eaway
    -0.96
    rostis
    -0.94
     começar
    -0.93
     sunrise
    -0.93
    えっ
    -0.92
    sund
    -0.91
     começ
    -0.90
     Resize
    -0.90
    solt
    -0.89
    POSITIVE LOGITS
     November
    1.32
     December
    1.07
    November
    1.05
     October
    1.02
     fall
    0.99
     noviembre
    0.98
    October
    0.98
     September
    0.91
     akhir
    0.88
    December
    0.87
    Activations Density 0.047%

    No Known Activations