© 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. 5717
    Prev
    Next
    INDEX
    Explanations

    LaTeX environments for math and lists

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

    This neuron fires on formal academic or mathematical terminology, especially phrases found in scholarly paper titles and citations.

    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
     italics
    0.61
     LEP
    0.59
    Zeit
    0.58
    StartY
    0.58
    পর্ব
    0.54
     россий
    0.51
    落ち
    0.51
    ({\
    0.51
    TEL
    0.51
    ほどの
    0.51
    POSITIVE LOGITS
    gathered
    0.73
    array
    0.64
    apple
    0.55
    arrays
    0.55
     Array
    0.54
     Luch
    0.53
    aligned
    0.52
     buys
    0.52
     Missions
    0.52
     selects
    0.52
    Activations Density 0.001%

    No Known Activations