© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 12890
    Prev
    Next
    INDEX
    Explanations

    code structures

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

    This neuron specializes in spotting literal‐delimiter tokens in code (e.g. the “%” array/string/vector markers and the matching quote, parenthesis, or bracket that begins a literal).

    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
     you
    -1.92
    琑
    -1.68
     šte
    -1.56
    ativement
    -1.55
     řešení
    -1.51
     opinión
    -1.49
     memudahkan
    -1.49
     najlep
    -1.48
     decisión
    -1.46
     menentukan
    -1.46
    POSITIVE LOGITS
    ).
    1.98
     were
    1.84
    Remarque
    1.66
    .”
    1.65
    ↵
    1.62
    Entrée
    1.61
    !
    1.55
    前回の
    1.54
    u
    1.52
    l
    1.48
    Activations Density 0.047%

    No Known Activations