© 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. 13974
    Prev
    Next
    INDEX
    Explanations

    crafts and burning

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

    This neuron strongly activates on the capitalized “How To …” tutorial‐style titles (e.g. “How To Clean A Wood Pellet Stove,” “How To Install A Wood Burning Stove In A Fireplace”).

    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
    冷蔵
    0.71
     neuropsych
    0.71
     SVP
    0.68
     Semiconductor
    0.66
     polypeptides
    0.64
    🍈
    0.64
     cryptographic
    0.64
     Lettuce
    0.63
     UIP
    0.61
     polypeptide
    0.60
    POSITIVE LOGITS
     burner
    2.14
     burners
    2.06
     fireplace
    2.02
    炉
    1.95
     flames
    1.92
     combustion
    1.91
     furnace
    1.91
    燃烧
    1.88
     burning
    1.86
     fireplaces
    1.86
    Activations Density 0.693%

    No Known Activations