© 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. 2785
    Prev
    Next
    INDEX
    Explanations

    java.util

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

    The neuron fires on numeric tokens—especially decimal numbers and version‐style numerals.

    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
    ั
    0.59
     человеком
    0.59
     splic
    0.59
     exosomes
    0.58
     सा
    0.57
     krijgen
    0.56
    redos
    0.56
    MethodBeat
    0.56
    ⓓ
    0.55
    舷
    0.55
    POSITIVE LOGITS
    lang
    1.09
    util
    0.96
    rmi
    0.95
    io
    0.90
    sql
    0.86
    awt
    0.81
     util
    0.81
    Lang
    0.80
     lang
    0.79
    UTIL
    0.79
    Activations Density 0.017%

    No Known Activations