© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 15849
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on numeric tokens—especially measurements, quantities, and statistical values—in the text.

    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
    यल
    0.52
    ेंसिल
    0.49
    ্লাহ
    0.48
    Ⲉ
    0.48
    它的
    0.46
    谏
    0.46
    勍
    0.46
    的一切
    0.46
    嚧
    0.46
    ᔭ
    0.45
    POSITIVE LOGITS
     each
    5.75
    each
    5.25
     EACH
    4.94
     Each
    4.88
    Each
    4.84
     каждого
    4.50
     каждой
    4.44
     каждая
    4.34
     fiecare
    4.28
    แต่ละ
    4.22
    Activations Density 1.162%

    No Known Activations