© 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. 9891
    Prev
    Next
    INDEX
    Explanations

    historically offer

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

    The neuron strongly activates on numeric tokens (multidigit numbers and decimal figures), i.e. quantitative 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
     ఇలా
    0.61
    ('');
    0.54
    一本
    0.52
    ясплат
    0.52
    мові
    0.52
    いち
    0.51
    sprintf
    0.51
    Примеча
    0.51
    こんな
    0.50
    ।''
    0.50
    POSITIVE LOGITS
     Trên
    0.56
     reis
    0.55
     leveraged
    0.55
     leverages
    0.55
    ǫ
    0.55
     dụng
    0.54
    dnn
    0.54
     dü
    0.54
    گاہ
    0.54
    abbr
    0.54
    Activations Density 0.001%

    No Known Activations