© 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. 1291
    Prev
    Next
    INDEX
    Explanations

    time durations

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

    This neuron activates on numeric expressions—whether digits or number words—especially counts and time spans (e.g. “few,” “six,” “weeks,” “1208.3043,” etc.).

    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
    any
    0.47
    ani
    0.46
     ゴールド
    0.45
    ujących
    0.45
    anguard
    0.44
     berbentuk
    0.44
    arios
    0.44
    乘
    0.44
    яр
    0.42
    chens
    0.42
    POSITIVE LOGITS
     months
    1.88
     weeks
    1.74
     month
    1.55
    months
    1.55
     Months
    1.51
     semanas
    1.48
     месяцев
    1.46
     miesią
    1.45
     maanden
    1.45
     Monaten
    1.45
    Activations Density 0.671%

    No Known Activations