© 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. 84266
    Prev
    Next
    INDEX
    Explanations

    data structures and flow

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

    This neuron detects programming identifiers or terms that include “data” (e.g. variable, function, or class names containing “Data”).

    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
     those
    -1.26
     produite
    -1.22
     oči
    -1.16
     fröh
    -1.13
     tém
    -1.09
    🧎
    -1.08
     fixé
    -1.06
     naive
    -1.05
    tagext
    -1.05
     détaillés
    -1.05
    POSITIVE LOGITS
    Its
    0.99
    芎
    0.98
    ↵↵
    0.97
    無法
    0.97
    ",
    0.95
    ',
    0.94
    года
    0.93
    For
    0.93
     appeals
    0.93
    Except
    0.93
    Activations Density 0.141%

    No Known Activations