© 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. 39480
    Prev
    Next
    INDEX
    Explanations

    programming keywords and types

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

    The neuron detects numeric literals (tokens containing digits) in code.

    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
     like
    -1.47
     there
    -1.39
     one
    -1.34
     their
    -1.33
    For
    -1.30
    There
    -1.29
     very
    -1.28
     more
    -1.26
     Before
    -1.24
    An
    -1.23
    POSITIVE LOGITS
    がありますが
    1.78
    があり
    1.70
    があるので
    1.45
    ఽ
    1.41
     모든
    1.36
    𐬹
    1.35
    various
    1.34
     vários
    1.33
    specifically
    1.33
    several
    1.32
    Activations Density 0.005%

    No Known Activations