© 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. 71711
    Prev
    Next
    INDEX
    Explanations

    cyclopedia entries and sections

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

    The neuron activates on standalone numeric tokens (e.g. years, list indices, counts) 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_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
     sosis
    -1.09
     parfü
    -1.09
    şen
    -1.05
    )";
    -1.05
    NFT
    -1.05
     cuma
    -1.05
     面包
    -1.05
     vermel
    -1.05
    NFTs
    -1.05
     pembun
    -1.04
    POSITIVE LOGITS
    <sup>
    1.43
     there
    1.25
    Pranala
    1.13
     who
    1.09
     noted
    1.01
     where
    1.00
     variar
    0.99
     see
    0.99
     discussion
    0.98
    <b>
    0.97
    Activations Density 0.077%

    No Known Activations