© 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. 8970
    Prev
    Next
    INDEX
    Explanations

    numbers after 5

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

    The neuron strongly activates on numerical tokens—especially floating‐point or large numeric values—i.e., it’s looking for numbers.

    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
     сек
    0.64
    性的
    0.63
    ूबी
    0.62
    మ్యాచ్
    0.61
    ंग्रेस
    0.61
    ਆ
    0.61
    ăpadă
    0.60
    جين
    0.59
    臭
    0.59
    ڑک
    0.58
    POSITIVE LOGITS
     деятельность
    0.63
    чков
    0.59
    Tron
    0.58
    ああ
    0.57
     means
    0.56
    मानस
    0.56
     устанавли
    0.55
     деятельности
    0.54
    мс
    0.54
     совершенно
    0.54
    Activations Density 0.124%

    No Known Activations