© 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. 13198
    Prev
    Next
    INDEX
    Explanations

    file paths or code definitions

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

    The neuron strongly activates on standalone numeric tokens—especially multi-digit figures or decimal numbers.

    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
    obt
    0.54
    etlen
    0.54
    पक
    0.52
    isseur
    0.52
     logements
    0.52
     儿
    0.52
    Ћ
    0.52
     čtyř
    0.50
     sine
    0.50
    DataContext
    0.50
    POSITIVE LOGITS
     ayudará
    0.68
    collegiate
    0.58
    注入
    0.57
     ചെയ്യുന്നത്
    0.57
     Helps
    0.57
    ിലുള്ള
    0.55
    돕
    0.55
     kas
    0.55
     Inject
    0.54
     एमपी
    0.53
    Activations Density 0.001%

    No Known Activations