© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 15362
    Prev
    Next
    INDEX
    Explanations

    string concatenation or placeholders

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

    The neuron activates on numeric literals (especially floating‐point constants) 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
     ancak
    -1.77
     حتی
    -1.74
     enlever
    -1.65
     ‌
    -1.62
    ținut
    -1.55
    NOON
    -1.52
     "..
    -1.49
     estekak
    -1.48
    骉
    -1.45
     Cependant
    -1.43
    POSITIVE LOGITS
     on
    1.84
     some
    1.77
     about
    1.71
     began
    1.67
     represents
    1.61
     became
    1.61
     chose
    1.61
     presents
    1.58
     starting
    1.56
     their
    1.55
    Activations Density 0.009%

    No Known Activations