© 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. 15640
    Prev
    Next
    INDEX
    Explanations

    The neuron selectively fires on numeric tokens representing large four-digit values (e.g. years or numbers above 2000).

    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
    y
    0.77
    ق
    0.74
     Avenida
    0.71
     rooms
    0.68
    j
    0.68
     archiv
    0.67
     MAPK
    0.66
    й
    0.66
     mRNA
    0.66
     Gigabit
    0.65
    POSITIVE LOGITS
     кто
    0.93
    ⓘ
    0.86
     несколько
    0.85
    が多く
    0.85
     mehrere
    0.84
     dotycz
    0.82
     zahlreiche
    0.80
    рые
    0.80
     помогут
    0.80
     некоторые
    0.80
    Activations Density 0.001%

    No Known Activations