© 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. 1752
    Prev
    Next
    INDEX
    Explanations

    This neuron detects numeric expressions and identifiers (numbers, dates, codes, or similar digit‐based tokens).

    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
     Pogis
    0.93
     bekannten
    0.88
    ٿ
    0.82
     ر
    0.82
    ُر
    0.82
    َہ
    0.82
    ۂ
    0.82
     ﺍﻟ
    0.81
    َق
    0.80
    ά
    0.80
    POSITIVE LOGITS
     via
    0.76
     because
    0.75
     due
    0.75
    .,
    0.74
    ();
    0.73
     despite
    0.73
     but
    0.71
    ​​​​
    0.68
     --
    0.65
     when
    0.65
    Activations Density 0.243%

    No Known Activations