© 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. 10-GEMMASCOPE-RES-131K
    7. 102186
    Prev
    Next
    INDEX
    Explanations

    UTF8 or non-ASCII characters

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

    numeric literals, especially the digit "1" and "2", in code or technical contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on numeric literal tokens (digits and numbers) in code or data.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    It
    -2.52
    Despite
    -2.39
    There
    -2.39
    5
    -2.36
    </h2>
    -2.25
    They
    -2.22
    0
    -2.22
    Unlike
    -2.19
    Because
    -2.14
    9
    -2.14
    POSITIVE LOGITS
     verschillende
    2.42
    犼
    2.41
     certaines
    2.39
    踯
    2.34
     plusieurs
    2.27
    麁
    2.23
     quintessential
    2.23
    袛
    2.20
     you
    2.17
     unbelievably
    2.16
    Activations Density 0.078%

    No Known Activations