© 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. 24-GEMMASCOPE-2-RES-16K
    7. 5169
    Prev
    Next
    INDEX
    Explanations

    quaternion ivector spacecode constrequirement definition

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

    This neuron primarily fires on numeric values—especially decimal numbers (e.g. prices, measurements, statistics).

    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_24_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
     ファン
    0.84
     즐
    0.80
     ある
    0.76
     pudi
    0.76
     増
    0.75
     Ад
    0.74
     เขียน
    0.74
     入れ
    0.73
     写
    0.73
    geben
    0.73
    POSITIVE LOGITS
    et
    0.82
     NA
    0.71
    نا
    0.71
    ع
    0.70
     craft
    0.68
    EM
    0.68
    నా
    0.68
     corridor
    0.68
    م
    0.68
    ÉT
    0.66
    Activations Density 0.000%

    No Known Activations