© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4689
    Prev
    Next
    INDEX
    Explanations

    describing product, brand, or medical information

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

    This neuron activates on standalone punctuation tokens—especially exclamation marks and colons—and other non-word symbols that serve as structural or formatting markers.

    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
    Box
    0.84
    Gets
    0.83
    onSubmit
    0.80
     adopts
    0.75
    형
    0.75
    STANDARD
    0.74
    습
    0.73
    컴
    0.73
    Lindsay
    0.72
    Bull
    0.71
    POSITIVE LOGITS
    的地方
    0.80
    煂
    0.77
    лив
    0.74
     barrios
    0.73
     murky
    0.72
     kön
    0.71
     ү
    0.71
     distanced
    0.71
     punkt
    0.71
     gebe
    0.70
    Activations Density 0.000%

    No Known Activations