© 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. 8840
    Prev
    Next
    INDEX
    Explanations

    The neuron reliably fires on the structured metadata fields of product listings—things like “sale since” dates, category paths, and similar listing‐specific tokens.

    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
    째
    0.75
     учет
    0.75
    ~
    0.75
    😏
    0.74
    ajax
    0.73
     NPE
    0.73
    $$\
    0.71
    ![](
    0.71
    breviations
    0.71
     ajax
    0.70
    POSITIVE LOGITS
     chyba
    0.74
     secondaires
    0.73
     крайней
    0.68
     Ashland
    0.68
     annan
    0.67
     grandes
    0.67
    Massachusetts
    0.66
     Lah
    0.66
     kawasan
    0.66
     serem
    0.65
    Activations Density 0.001%

    No Known Activations