© 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. 6640
    Prev
    Next
    INDEX
    Explanations

    code decorators and annotations

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

    The neuron fires on code‐style annotation and metadata identifiers (e.g. annotations like @NonNull, @Delegate, @WithType, Override, Deprecated, read‐only, version tags, Definition, Runtime, etc.).

    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
    ा
    0.64
    UT
    0.53
    a
    0.53
    ্য
    0.52
    UA
    0.51
    UB
    0.50
    ừ
    0.47
    UR
    0.47
    არე
    0.47
    ıya
    0.46
    POSITIVE LOGITS
    fdPar
    0.67
     nomencl
    0.67
    trifluoromethyl
    0.64
     motore
    0.62
     interchangeably
    0.62
    のために
    0.60
     (†
    0.60
     utilisés
    0.59
     cuáles
    0.59
    世代
    0.59
    Activations Density 0.000%

    No Known Activations