© 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. 9227
    Prev
    Next
    INDEX
    Explanations

    frequency and time

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

    The neuron primarily detects numeric tokens (digits and time values).

    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
    hém
    0.55
    holt
    0.51
    clampsia
    0.50
    HMS
    0.49
    holtz
    0.48
     tetanus
    0.48
     அலங்க
    0.48
    枥
    0.48
     demonic
    0.47
     osl
    0.47
    POSITIVE LOGITS
     Mondays
    2.44
     weekdays
    2.42
     Tuesdays
    2.38
     Monday
    2.30
    Monday
    2.16
     weekday
    2.13
     Tuesday
    2.06
     Wednesdays
    2.03
     monday
    1.98
     सोमवार
    1.96
    Activations Density 0.563%

    No Known Activations