© 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. 19540
    Prev
    Next
    INDEX
    Explanations

    ferrous salts, Folio, emulate, semaphore, toDate, emulation

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

    The neuron specializes in spotting uncommon technical or branded terms—e.g. chemical/material names (like “ferrous”), product or model names (“Folio,” “Semaphore”), and similar domain-specific jargon.

    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
    其妙
    -2.53
    垸
    -2.52
     alebo
    -2.48
     juſ
    -2.45
    镡
    -2.45
    恸
    -2.42
     нам
    -2.42
     ženy
    -2.38
     августа
    -2.33
    tômes
    -2.31
    POSITIVE LOGITS
    '
    3.39
    This
    2.73
    h
    2.72
            
    2.67
    ↵
    2.53
        
    2.50
     высокого
    2.48
       
    2.48
         
    2.33
     selben
    2.25
    Activations Density 0.012%

    No Known Activations