© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 12459
    Prev
    Next
    INDEX
    Explanations

    code delimiters ({}:,.)

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

    The neuron strongly activates on standalone numeric literals (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_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
     for
    -1.64
     those
    -1.50
     by
    -1.34
    :
    -1.27
    "}}>
    -1.27
    \\
    -1.27
     the
    -1.27
    ⌉
    -1.25
    ˗
    -1.25
               
    -1.25
    POSITIVE LOGITS
    怎么回事
    1.66
     rafraî
    1.64
     refroid
    1.63
     verrou
    1.63
     beneficiario
    1.62
     banyo
    1.62
     récompense
    1.61
     einzige
    1.59
     uniquement
    1.53
    Tentang
    1.52
    Activations Density 0.181%

    No Known Activations