© 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. 2681
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and greek letters

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

    The neuron strongly activates on HTML table‐related tag tokens (e.g. “td” and variations), i.e. it detects table cell markup.

    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
    Invoice
    0.78
    Treasurer
    0.76
    ຕ້ອງການ
    0.75
     akadem
    0.74
     republika
    0.74
    ⁩
    0.73
     había
    0.73
    businesses
    0.73
     बिजनेस
    0.72
     eröffnet
    0.72
    POSITIVE LOGITS
     =
    1.17
     −
    1.15
     ±
    1.06
    ₁+
    1.04
     =-
    1.02
     +
    1.02
     ∈
    0.98
     x
    0.94
     -=
    0.93
     ≤
    0.92
    Activations Density 0.122%

    No Known Activations