© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7014
    Prev
    Next
    INDEX
    Explanations

    decades (19xxs)

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

    The neuron is specialized in spotting numeric tokens (years, dates, measurements, block numbers, and other digit strings).

    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
    身份
    0.51
    勻
    0.49
    鎏
    0.49
    あい
    0.48
    㭚
    0.47
     simple
    0.45
    とり
    0.44
    閂
    0.44
    strftime
    0.44
    ቤት
    0.44
    POSITIVE LOGITS
     thirties
    0.85
     twenties
    0.84
     eighties
    0.77
    reeks
    0.75
     nineties
    0.75
     fifties
    0.74
     seventies
    0.74
     Hundreds
    0.73
    Hundreds
    0.73
     ratus
    0.73
    Activations Density 0.162%

    No Known Activations