© 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. 12972
    Prev
    Next
    INDEX
    Explanations

    step-by-step, multiple times, more than

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

    The neuron specializes in picking out numerical or quantitative expressions—tokens that convey counts, measures, or magnitudes.

    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
     års
    0.62
     главной
    0.59
    عة
    0.58
     ThemeData
    0.58
     mey
    0.58
     Aix
    0.58
     Mysterious
    0.56
    圴
    0.56
     Approach
    0.55
     Такой
    0.55
    POSITIVE LOGITS
     way
    1.12
    ologically
    0.96
    ographically
    0.95
    way
    0.94
    heartedly
    0.94
     façon
    0.88
    ically
    0.84
    lessly
    0.84
    clusively
    0.83
    operatively
    0.82
    Activations Density 0.732%

    No Known Activations