© 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. 128657
    Prev
    Next
    INDEX
    Explanations

    art contexts

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

    The neuron strongly activates on tokens that occur in section headings or title‐like metadata (capitalized words, numbers, and punctuation/symbol markers used in headings).

    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
    ’
    -1.95
    ámide
    -1.47
    ｼ
    -1.45
     for
    -1.44
    ଃ
    -1.43
    ’，
    -1.41
     avendo
    -1.34
    یت
    -1.34
    ll
    -1.34
    痦
    -1.33
    POSITIVE LOGITS
     it
    1.53
     principali
    1.52
     wonderful
    1.51
     referido
    1.48
    ↵↵
    1.45
     contemporanea
    1.45
    ?')
    1.42
    ');
    1.41
     vanta
    1.37
     refugiados
    1.35
    Activations Density 0.050%

    No Known Activations