© 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. 98384
    Prev
    Next
    INDEX
    Explanations

    names in email headers

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

    The neuron detects email header and routing metadata (e.g. “From:”, “Forwarded by”, email addresses, timestamps, slash‐delimited fields).

    oai_token-act-pair · o4-miniTriggered 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
     make
    -1.04
     during
    -1.03
     Felsen
    -1.02
     Instagram
    -1.00
     may
    -0.99
     with
    -0.99
     perlu
    -0.98
     pentru
    -0.98
     get
    -0.97
     less
    -0.97
    POSITIVE LOGITS
     считают
    1.23
     vannak
    1.22
    archiwizowane
    1.20
     メッセージ
    1.20
    würdig
    1.20
    ută
    1.20
     incendios
    1.17
    especially
    1.17
    particularly
    1.15
    Especially
    1.13
    Activations Density 0.107%

    No Known Activations