© 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. 130503
    Prev
    Next
    INDEX
    Explanations

    Antibiotic, detecting, James

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

    special characters or punctuation marks in technical and scientific text.

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

    This neuron detects words formed with negative prefixes (e.g., “un-…” or “non-…”).

    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
    
    -1.63
    ⢇
    -1.41
     irmã
    -1.32
    
    -1.30
    strSql
    -1.29
     quintessential
    -1.29
    方は
    -1.27
     sembla
    -1.26
     pivotal
    -1.24
    蟊
    -1.24
    POSITIVE LOGITS
     that
    1.93
    there
    1.71
     at
    1.51
    してみて
    1.48
     but
    1.48
    There
    1.48
    هاز
    1.37
    as
    1.34
    稅
    1.33
    After
    1.32
    Activations Density 0.000%

    No Known Activations