© 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. 5181
    Prev
    Next
    INDEX
    Explanations

    latex and code punctuation

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

    The neuron is most strongly activated by social‐media style metadata and URL formatting tokens—e.g. “likes” counts and the “/” characters in web addresses.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    lett
    0.64
    lashes
    0.57
     Posteriormente
    0.54
    ంచు
    0.54
    ertes
    0.53
    passing
    0.53
    歹
    0.53
    textbf
    0.53
    =['
    0.53
    buried
    0.52
    POSITIVE LOGITS
     реализова
    0.60
     Shutterstock
    0.59
     diput
    0.59
     quarts
    0.58
     SDR
    0.57
     jScrollPane
    0.57
     quarter
    0.56
     kilo
    0.56
    新聞
    0.55
     SOD
    0.55
    Activations Density 0.003%

    No Known Activations