© 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. 22-GEMMASCOPE-RES-131K
    7. 36444
    Prev
    Next
    INDEX
    Explanations

    enron fm emails

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

    This neuron detects structured metadata header lines composed of uppercase classification labels, routing codes, dates, and other alphanumeric tags.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    email header formatting and metadata (including timestamps, sender/recipient information, and message routing details).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ׇ
    -1.46
     for
    -1.41
     Often
    -1.30
     osv
    -1.26
     FOR
    -1.23
     people
    -1.21
     блюдо
    -1.20
     Belajar
    -1.19
     Banyak
    -1.17
     because
    -1.16
    POSITIVE LOGITS
    ּוֹ
    1.45
    enablog
    1.45
     trám
    1.36
    ִּ
    1.35
     xay
    1.34
     ninh
    1.34
    ensitive
    1.34
    憷
    1.33
    ְּ
    1.31
     rác
    1.27
    Activations Density 0.076%

    No Known Activations