© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 72072
    Prev
    Next
    INDEX
    Explanations

    computer hardware and systems

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

    The neuron detects prominent domain-specific topic nouns—that is, words naming central subjects or concepts in the text.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     refroid
    -1.77
     réfrig
    -1.70
    閪
    -1.55
     rafraî
    -1.53
     véhic
    -1.52
     peluche
    -1.52
     nettoy
    -1.47
     éta
    -1.47
     caricatura
    -1.47
     dalamnya
    -1.46
    POSITIVE LOGITS
     to
    2.80
     where
    2.06
     into
    1.98
     among
    1.95
     including
    1.84
     along
    1.83
     for
    1.81
     when
    1.80
     through
    1.67
     following
    1.66
    Activations Density 0.251%

    No Known Activations