© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13952
    Prev
    Next
    INDEX
    Explanations

    months and days of week

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

    The neuron is detecting numeric tokens—especially dates, years, times, and HTML‐encoded numbers.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     supporter
    0.59
    supporters
    0.57
     supporters
    0.55
    δό
    0.54
    בל
    0.53
    support
    0.51
    umed
    0.51
    섰
    0.51
    blank
    0.50
    ций
    0.50
    POSITIVE LOGITS
    することで
    0.64
     Feb
    0.61
     Exploring
    0.60
     ·
    0.59
     exploring
    0.58
     Andrew
    0.57
     Minn
    0.57
    Exploring
    0.56
     Aug
    0.56
     Hilary
    0.55
    Activations Density 0.097%

    No Known Activations