© 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. 20501
    Prev
    Next
    INDEX
    Explanations

    court filings

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

    This neuron fires on spaced‐out uppercase letter sequences as used in big all-caps section headers (e.g. “F O O L W A T C H”).

    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
    0
    -2.56
    Surprisingly
    -2.17
    5
    -2.16
    With
    -2.16
    Even
    -2.05
    Despite
    -2.03
    4
    -2.03
    From
    -2.03
    1
    -2.02
    3
    -1.98
    POSITIVE LOGITS
     to
    2.05
     exploración
    2.00
     alcuni
    1.96
     !!!!
    1.95
     Anſ
    1.91
    蛲
    1.88
     厂家
    1.86
    鹋
    1.81
     semelh
    1.74
     filosofía
    1.74
    Activations Density 0.019%

    No Known Activations