© 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. 2242
    Prev
    Next
    INDEX
    Explanations

    trained by

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

    The neuron is triggered by large numeric quantities (multi-digit numbers) denoting magnitudes such as monetary amounts or 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_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
     Küçük
    0.64
     peeling
    0.63
     पड़ती
    0.62
     forearm
    0.61
     paypal
    0.59
     خفی
    0.59
     küçük
    0.59
    珢
    0.59
     manus
    0.58
     motel
    0.58
    POSITIVE LOGITS
     billions
    3.03
     billion
    2.83
     миллиар
    2.66
    billion
    2.45
     milliards
    2.44
     مليار
    2.41
     Milliarden
    2.39
     miliardi
    2.39
     bilhões
    2.38
     trillions
    2.36
    Activations Density 0.455%

    No Known Activations