© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8997
    Prev
    Next
    INDEX
    Explanations

    options and categories

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

    The neuron primarily detects mentions of user‐interface controls—especially drop-down list widgets.

    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_31_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
     Commentaires
    0.57
    쩡
    0.56
    RecordPtr
    0.52
     ठे
    0.52
     ক্ষতির
    0.51
     tracer
    0.50
     張
    0.50
     Ztg
    0.50
    ণিত
    0.50
     *}$
    0.49
    POSITIVE LOGITS
    dropdown
    1.30
     choices
    1.24
    DropDownList
    1.23
    Dropdown
    1.20
    DropDown
    1.19
    choices
    1.19
    Choices
    1.17
    SelectList
    1.16
    下拉
    1.14
     dropdown
    1.12
    Activations Density 0.059%

    No Known Activations