© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6493
    Prev
    Next
    INDEX
    Explanations

    code blocks and classes

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

    The neuron fires on technical API identifiers—especially CamelCase class and type names in code or documentation.

    oai_token-act-pair · o4-miniTriggered 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
    :(
    0.61
    :"
    0.58
     ОТ
    0.58
    %:
    0.55
    ?:
    0.55
    :,
    0.54
     НЕ
    0.54
    :%
    0.54
    Gradle
    0.54
    :
    0.54
    POSITIVE LOGITS
     model
    0.60
     obiekt
    0.60
     slander
    0.56
     palace
    0.54
     Mereka
    0.54
     mengaku
    0.53
     nowy
    0.51
     speculator
    0.50
     tann
    0.50
    知県
    0.50
    Activations Density 0.008%

    No Known Activations