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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13280
    Prev
    Next
    INDEX
    Explanations

    subject followed by verb

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

    This neuron detects document-structure markers such as titles, headings, separators, and bylines.

    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
     He
    -1.38
     quedaba
    -1.19
    ,
    -1.18
     Fantastic
    -1.13
    Who
    -1.13
    分けて
    -1.09
     Hochsch
    -1.08
     Gold
    -1.06
    right
    -1.06
     set
    -1.06
    POSITIVE LOGITS
     segala
    1.24
     first
    1.21
     mercurial
    1.13
     REDUC
    1.13
     riuscito
    1.12
    ihnachts
    1.09
    riever
    1.09
     吊
    1.08
     restau
    1.06
    Crap
    1.06
    Activations Density 0.235%

    No Known Activations