© 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. 130838
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on code identifiers—names of variables, functions, types, etc.—in programming snippets.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    code definition or variable name definition

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     mereka
    -0.85
    Wound
    -0.80
    iedler
    -0.78
    ḍ
    -0.78
     well
    -0.77
    kenalkan
    -0.75
     moderno
    -0.75
    rierung
    -0.75
    genen
    -0.75
    mission
    -0.74
    POSITIVE LOGITS
     Vorteile
    0.88
    ᅟ
    0.88
    >::
    0.88
     زیادی
    0.88
     empfangen
    0.88
     chùa
    0.83
    إ
    0.83
    usor
    0.81
    espa
    0.80
    ৮
    0.79
    Activations Density 0.014%

    No Known Activations