© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 73880
    Prev
    Next
    INDEX
    Explanations

    case reports and discussions

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

    patterns where a person shares, presents, describes, discusses, or reviews information in a report, article, or publication.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on reporting-style verbs (like shares, exposes, discusses, announces, presents, describes, reviews) that introduce or summarize content.

    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
     in
    -2.28
     here
    -1.72
     there
    -1.54
     then
    -1.50
     at
    -1.49
     it
    -1.36
     throughout
    -1.30
     when
    -1.30
     []:
    -1.23
     thereafter
    -1.20
    POSITIVE LOGITS
     två
    1.47
     how
    1.23
    yaitu
    1.17
    AKERS
    1.14
     dáms
    1.14
     fyra
    1.14
    archiwizowane
    1.13
    了一眼
    1.11
     ;
    1.11
    んびり
    1.10
    Activations Density 0.189%

    No Known Activations