© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5630
    Prev
    Next
    INDEX
    Explanations

    starting or being

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

    The neuron responds to code identifier tokens (e.g., variable, function, or field names).

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

    variable names in programming code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    programming-related identifiers and code tokens, especially variable, method, field, and file/path names in technical text and code snippets.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enni
    0.69
     Who
    0.66
     निवड
    0.66
     वैशिष्ट
    0.66
    ?](
    0.65
     पादुकोण
    0.64
     hickory
    0.64
     रिप्रे
    0.64
    ǒu
    0.63
    ”?
    0.63
    POSITIVE LOGITS
     =
    1.04
     был
    0.68
    _=
    0.68
    は
    0.68
    氏は
    0.67
    setKeep
    0.65
     wrote
    0.62
    年は
    0.61
    は大
    0.60
    newEvent
    0.60
    Activations Density 0.001%

    No Known Activations