© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 5263
    Prev
    Next
    INDEX
    Explanations

    historical significance, basic ideas, core concepts

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

    The neuron fires strongly on dense, formal academic or mathematical discourse—i.e. technical jargon, definitions, and formula-heavy segments typical of research writing.

    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
    itosan
    0.80
    ));//
    0.72
     foodservice
    0.70
    ˃
    0.70
    DIN
    0.69
    korzyst
    0.68
    ᵒ
    0.68
    Ɵ
    0.68
    crawler
    0.67
     crawler
    0.66
    POSITIVE LOGITS
     perturbative
    1.09
     conjectured
    1.00
     parametrized
    1.00
     arxiv
    0.99
     motivic
    0.98
     extremal
    0.98
     simplicial
    0.95
    推广
    0.93
     generality
    0.88
    ≳
    0.88
    Activations Density 0.132%

    No Known Activations