© 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. 7824
    Prev
    Next
    INDEX
    Explanations

    sw or sw followed by specific words

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

    The neuron strongly activates on occurrences of the subword “Sw” (e.g. in names like Swartwout, Swanson, Switzler).

    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
    unday
    0.56
    ourse
    0.55
    ὸ
    0.52
    इसे
    0.51
    𝓪
    0.50
     semicircle
    0.50
    llas
    0.49
     Tatsache
    0.48
    備
    0.48
    rame
    0.47
    POSITIVE LOGITS
    Sw
    1.18
     Sw
    1.12
     SW
    1.05
     sw
    1.05
    SW
    1.01
     Swagger
    0.92
     Swe
    0.90
     SWE
    0.90
    SWE
    0.87
     Swans
    0.86
    Activations Density 0.180%

    No Known Activations