© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 65654
    Prev
    Next
    INDEX
    Explanations

    confidentiality, employment, contracts, competition law

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

    The neuron detects mentions of confidential or proprietary information—especially terms like “trade secrets,” “confidential information,” and related legal secrecy language.

    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
     batterie
    -1.44
    <td>
    -1.25
     alimente
    -1.23
     vei
    -1.23
     hiver
    -1.23
     vider
    -1.22
     apparence
    -1.20
    慙
    -1.20
     Mockito
    -1.19
    :''
    -1.16
    POSITIVE LOGITS
     ALWAYS
    1.34
     If
    1.27
     as
    1.25
     solely
    1.23
     Because
    1.21
     their
    1.21
     Even
    1.20
     at
    1.20
     purposefully
    1.19
     that
    1.17
    Activations Density 0.045%

    No Known Activations