© 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. 7652
    Prev
    Next
    INDEX
    Explanations

    street context

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

    The neuron strongly activates on the token “Street,” i.e. it detects occurrences of the word “Street.”

    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
    ン
    0.65
    richTextBox
    0.64
    powers
    0.62
    리
    0.57
    키
    0.57
    وج
    0.57
    FileSize
    0.56
    내
    0.56
     따라서
    0.55
    중
    0.55
    POSITIVE LOGITS
    车站
    0.65
     Gotham
    0.64
    inati
    0.59
    ૂત
    0.58
     Bahnhof
    0.57
     STATION
    0.57
     WALK
    0.57
    車站
    0.57
     begon
    0.55
     Zeke
    0.55
    Activations Density 0.010%

    No Known Activations