© 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. 22-GEMMASCOPE-RES-131K
    7. 24974
    Prev
    Next
    INDEX
    Explanations

    abstract nouns and concepts

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

    This neuron activates on salient content words—especially nouns and abstract concepts—marking them as key topics or themes 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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     bluza
    -1.43
     variaciones
    -1.38
     sonriente
    -1.36
     également
    -1.33
    為に
    -1.31
     wzglę
    -1.27
     probablement
    -1.27
     récol
    -1.26
     probablemente
    -1.25
     delikat
    -1.25
    POSITIVE LOGITS
     is
    2.55
    が
    1.65
     must
    1.59
     cannot
    1.50
    :
    1.48
     itself
    1.47
     should
    1.34
     in
    1.33
     not
    1.32
     =
    1.29
    Activations Density 0.027%

    No Known Activations