© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10328
    Prev
    Next
    INDEX
    Explanations

    alphanumeric strings

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

    The neuron is detecting long sequences of digits and mixed alphanumeric codes (e.g., transaction IDs, barcodes, account numbers, and similar structured identifiers).

    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
     sareng
    0.31
     Overflow
    0.30
     thumbnailUrl
    0.29
     Dreaming
    0.29
     تضم
    0.28
     قطعة
    0.28
     ThemeOverlay
    0.28
    ꩡ
    0.27
     Lyrics
    0.27
     dovrebbe
    0.27
    POSITIVE LOGITS
    MDA
    0.30
    b
    0.30
    V
    0.29
    
    0.29
    Сер
    0.29
    prom
    0.29
    Band
    0.29
    D
    0.29
    Из
    0.29
    borne
    0.28
    Activations Density 0.086%

    No Known Activations