© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 2400
    Prev
    Next
    INDEX
    Explanations

    late or today

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

    The main thing this neuron does is detect numerical tokens, especially those used in dates, times, and other numeric references.

    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
     aldrig
    0.57
     coatings
    0.55
    梖
    0.53
    ുകളും
    0.53
    نمية
    0.53
     maddenin
    0.53
     silam
    0.52
    লিখিত
    0.51
     listop
    0.51
     दिवशी
    0.50
    POSITIVE LOGITS
    9
    0.86
    7
    0.81
    4
    0.80
    8
    0.79
    6
    0.78
    3
    0.73
    5
    0.71
    ۹
    0.69
    
    0.66
    ٩
    0.65
    Activations Density 0.692%

    No Known Activations