© 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. 2934
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is detect numeric tokens (digits and numbers) in the text.

    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
     širo
    0.64
     Shor
    0.63
     Š
    0.62
     MCS
    0.57
     atract
    0.57
     Verkehrs
    0.57
     SCS
    0.55
     Composite
    0.54
    नावश्यक
    0.54
     attract
    0.54
    POSITIVE LOGITS
    9
    0.90
    7
    0.90
    abcdef
    0.89
    abc
    0.83
    8
    0.83
    abcd
    0.81
    6
    0.81
    xyz
    0.79
    zzz
    0.77
    asd
    0.77
    Activations Density 0.099%

    No Known Activations