© 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. 118874
    Prev
    Next
    INDEX
    Explanations

    ectopic or out-of-place tissues

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

    words that are out of place or context, like "ectopic," "anachronism," or "out of place."

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron is activated most strongly by longer, low‐frequency content words—that is, uncommon or multi-syllabic nouns and technical terms.

    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
    You
    -1.30
    a
    -1.30
     expérimentés
    -1.25
    实际
    -1.23
     Verlust
    -1.17
     anstatt
    -1.12
     garantizar
    -1.11
    LessThan
    -1.11
     définiti
    -1.11
     wzgl
    -1.10
    POSITIVE LOGITS
     in
    1.54
    ± 
    1.35
     here
    1.26
     ملع
    1.26
    ૬
    1.24
     Therm
    1.23
     있는
    1.21
    ulsión
    1.20
     opérations
    1.18
    zumaki
    1.17
    Activations Density 0.042%

    No Known Activations