© 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. 6923
    Prev
    Next
    INDEX
    Explanations

    life, health, and medicine

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

    The neuron strongly activates on numeric tokens—especially floating-point numbers or measurements—i.e. decimal or large numerical values.

    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
      
    0.73
     which
    0.71
     которой
    0.71
    which
    0.71
     welches
    0.66
     والتي
    0.63
    Ce
    0.63
     mre
    0.62
    <0xC2>
    0.58
    Which
    0.57
    POSITIVE LOGITS
     सफाया
    0.67
     그다음에
    0.67
     опять
    0.66
    ଣ୍
    0.66
    ேத்க
    0.63
     stockholder
    0.60
    VCTarget
    0.59
    lardı
    0.58
     গিয়েছিল
    0.57
     asymmetries
    0.57
    Activations Density 0.012%

    No Known Activations