© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5929
    Prev
    Next
    INDEX
    Explanations

    converting fractions and decimals

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

    The neuron fires on domain‐specific jargon and proper‐name tokens—e.g. currency codes, product or company names, technical labels, and section‐heading words—that mark shifts into specialized topics.

    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
     ออก
    0.81
     зарегистри
    0.75
     ঢাকায়
    0.71
     למ
    0.71
     DET
    0.69
     צו
    0.68
     カスタ
    0.68
     בא
    0.68
     לק
    0.68
     ঢাকায়
    0.68
    POSITIVE LOGITS
    我們會
    0.78
    ractions
    0.77
     mathematic
    0.77
    小数
    0.74
    我们
    0.73
     mathematically
    0.71
    整数
    0.70
    我们会
    0.70
     Rashford
    0.69
     हाम्रो
    0.69
    Activations Density 0.002%

    No Known Activations