© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6643
    Prev
    Next
    INDEX
    Explanations

    search in

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

    The neuron specializes in spotting words that denote the act of seeking or looking (e.g. “search,” “tracking,” “looking”).

    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
    塑造
    0.58
    軖
    0.57
     deformed
    0.55
     बिगड़
    0.54
    影响
    0.53
    娇
    0.52
     rejuven
    0.51
    影響
    0.51
     corrective
    0.50
     Ə
    0.50
    POSITIVE LOGITS
     search
    3.08
     searches
    2.91
    搜索
    2.86
    search
    2.84
     searching
    2.83
     Search
    2.77
     SEARCH
    2.73
    searching
    2.72
     searched
    2.70
     Searching
    2.69
    Activations Density 0.687%

    No Known Activations