© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 42680
    Prev
    Next
    INDEX
    Explanations

    new season's books

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

    The neuron strongly fires on words that appear in headings, titles, or other prominently formatted “headline” text (e.g. section titles, ad‐style banners, or bullet‐list starters).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     decoradas
    -1.47
     Küste
    -1.38
     seguras
    -1.32
     Hauptstadt
    -1.29
    してた
    -1.29
     rápidas
    -1.28
    escola
    -1.23
     famosas
    -1.23
     fotó
    -1.22
     preciosas
    -1.22
    POSITIVE LOGITS
     Then
    1.78
     also
    1.75
     instead
    1.55
     will
    1.51
     ducato
    1.47
     then
    1.40
     aussitôt
    1.38
     icona
    1.38
     Instead
    1.38
     our
    1.37
    Activations Density 0.010%

    No Known Activations