© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130272
    Prev
    Next
    INDEX
    Explanations

    The neuron primarily detects standalone numerals (e.g. years, dates, and other number tokens).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    years and numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    (?:
    -1.02
     kend
    -0.98
     antaranya
    -0.98
     toer
    -0.97
     hulle
    -0.97
    ὕ
    -0.95
     ƙ
    -0.95
     plati
    -0.94
     feit
    -0.93
     zor
    -0.92
    POSITIVE LOGITS
     leaves
    1.05
     stirred
    0.99
     leaving
    0.98
    Instead
    0.95
    Are
    0.95
     stepped
    0.94
     inviting
    0.94
     hark
    0.93
     类
    0.93
     why
    0.92
    Activations Density 0.001%

    No Known Activations