© 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. 5785
    Prev
    Next
    INDEX
    Explanations

    MiFID, qi Investment, Wireshark

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

    The neuron consistently lights up on capitalized, uncommon tokens—i.e. proper names, titles, and acronyms indicating named entities.

    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
    েই
    1.07
    ました
    1.06
    ます
    1.04
    ni
    1.02
    ्स
    0.99
    از
    0.99
    ため
    0.98
    ন
    0.95
    ם
    0.93
    نی
    0.90
    POSITIVE LOGITS
    깐
    1.20
    ٰ
    1.04
    ñ
    0.95
    ğinde
    0.95
    ć
    0.95
    ppo
    0.94
    ながら
    0.92
    деся
    0.92
    ště
    0.91
    ście
    0.89
    Activations Density 1.241%

    No Known Activations