© 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. 9683
    Prev
    Next
    INDEX
    Explanations

    code constants and final variables

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

    The neuron fires on language‐level constant/immutable declarations—e.g. the `final` or `const` keywords introducing fixed-value variables.

    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
     nadru
    -1.74
     histó
    -1.70
    鐐
    -1.69
     salonu
    -1.67
     Diputados
    -1.67
     וכן
    -1.60
    Waktu
    -1.59
     bluz
    -1.58
     kancel
    -1.58
    dä
    -1.57
    POSITIVE LOGITS
    E
    2.03
    '
    1.98
    _
    1.98
     is
    1.91
    It
    1.88
    –
    1.87
    I
    1.84
    A
    1.84
    *
    1.84
    1.83
    Activations Density 0.002%

    No Known Activations