© 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. 12619
    Prev
    Next
    INDEX
    Explanations

    Regulations and requirements

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

    The neuron activates on numerical tokens—quantities, statistics, measurements, or other number‐related 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
    ๒
    0.66
    Ｄ
    0.60
    𝗗
    0.58
     Decreto
    0.57
    ๑
    0.57
    僭
    0.57
    legal
    0.56
    ugd
    0.56
     Règlement
    0.56
    ukat
    0.56
    POSITIVE LOGITS
     चीजों
    0.71
    됬
    0.71
     সেটা
    0.70
    像是
    0.67
    কোনও
    0.67
     ভ্রম
    0.65
     ciertas
    0.63
     जरुरी
    0.63
     ciertos
    0.62
     दुनियाभर
    0.62
    Activations Density 0.154%

    No Known Activations