© 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. 98540
    Prev
    Next
    INDEX
    Explanations

    months and numbers

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

    The neuron fires on numeric tokens (digits and numbers) appearing in technical specs, dates, times, and document‐metadata.

    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
     alkoholu
    -1.08
     lumineux
    -0.98
     parfaitement
    -0.91
    tų
    -0.91
     sadece
    -0.90
    ሲ
    -0.88
    tatuajes
    -0.88
    ​​​
    -0.88
     skola
    -0.87
     uniquement
    -0.87
    POSITIVE LOGITS
     will
    1.11
     with
    1.00
     there
    0.97
     additional
    0.96
    擒
    0.91
     and
    0.90
     should
    0.90
    の人は
    0.90
     on
    0.89
    ↵↵↵↵
    0.87
    Activations Density 0.068%

    No Known Activations