© 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. 69158
    Prev
    Next
    INDEX
    Explanations

    medical conditions, places, and proper nouns

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

    non-English words, particularly from Asian and Middle Eastern languages.

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

    This neuron fires on uncommon subword fragments—especially those from rare or technical terms, non-English proper names, medical jargon, or code/hex sequences.

    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
     (>
    -1.97
     of
    -1.78
     any
    -1.73
     be
    -1.68
     даже
    -1.60
     $\
    -1.56
    𓍊
    -1.55
    )
    -1.53
     любви
    -1.52
     любой
    -1.51
    POSITIVE LOGITS
     ';
    
    1.90
     ſuch
    1.88
    erdings
    1.88
    汌
    1.83
    泚
    1.82
     駅前
    1.77
     emplois
    1.75
    berra
    1.75
    取决
    1.75
     againſt
    1.73
    Activations Density 0.101%

    No Known Activations