© 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. 34-GEMMASCOPE-RES-131K
    7. 49910
    Prev
    Next
    INDEX
    Explanations

    column values or identifiers

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

    The neuron fires on standalone numeric literals (digit‐only tokens) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     even
    -1.20
     without
    -1.20
     any
    -1.09
     so
    -1.09
    dientes
    -1.06
     vann
    -1.03
     satt
    -1.00
     from
    -0.99
     there
    -0.98
     their
    -0.98
    POSITIVE LOGITS
    をご覧
    1.19
    這裡
    1.06
     wenigstens
    1.05
    bouton
    1.04
     sommige
    1.04
     blossoming
    1.03
     begrenzt
    1.03
     möglicherweise
    1.03
    这两
    1.02
    création
    1.02
    Activations Density 0.012%

    No Known Activations