© 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. 11350
    Prev
    Next
    INDEX
    Explanations

    years ago

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

    The neuron detects date/time expressions, especially four-digit year references (e.g. “In 1997,” “November 2018,” “August 2021”).

    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
     culminating
    1.11
     onwards
    1.09
     but
    1.05
     coinciding
    1.05
    ;
    1.02
     nhưng
    1.01
     when
    0.98
     with
    0.97
     —
    0.97
    .[
    0.95
    POSITIVE LOGITS
    わからない
    0.78
    长时间
    0.77
    今回の
    0.70
    先日
    0.69
    めます
    0.68
     сьогодні
    0.68
     kürz
    0.68
    长的
    0.67
     ayer
    0.67
     আজকাল
    0.67
    Activations Density 0.106%

    No Known Activations