© 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. 2958
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on the “m” token in time stamps (as in the “a.m.” or “p.m.” markers).

    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
    先端
    0.61
    нося
    0.58
     भाइयों
    0.55
    ţa
    0.55
     escalating
    0.55
    ontium
    0.55
    싼
    0.55
     fluorine
    0.54
    zeichnet
    0.54
    হিত
    0.54
    POSITIVE LOGITS
     ✨
    0.55
    JWT
    0.55
    ↵↵↵↵
    0.55
     CST
    0.55
    ↵↵↵↵↵
    0.54
     vak
    0.54
    >());
    0.53
     imassa
    0.52
    🐙
    0.52
    ↵↵↵↵↵↵
    0.52
    Activations Density 0.000%

    No Known Activations