© 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. 5049
    Prev
    Next
    INDEX
    Explanations

    This neuron fires most strongly on standalone topic nouns or section‐heading words (single substantive terms that label the subject of a passage).

    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
     firstly
    0.66
    并将
    0.65
    1
    0.65
    0.65
    Already
    0.63
     Already
    0.61
    并且
    0.61
    หรือไม่
    0.60
    时间和
    0.59
    并
    0.59
    POSITIVE LOGITS
    はその
    1.21
     is
    1.16
     becomes
    1.13
     blijft
    1.13
     یې
    1.12
     thereof
    1.09
     wordt
    1.05
     was
    1.05
     has
    1.05
     thereunder
    1.04
    Activations Density 0.588%

    No Known Activations