© 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. 124012
    Prev
    Next
    INDEX
    Explanations

    linking importing to uncompressed files

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

    The neuron is sensitive to content words (especially nouns) together with their linking prepositions in descriptive or title-style phrases—e.g. technical file‐import steps, place-of-location constructions, formal names (“Member States within the Council and the Commission”), configuration settings (“on your home directory”), and similar noun‐preposition 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
     wasn
    -1.65
     doesn
    -1.62
     I
    -1.51
     didn
    -1.50
    愜
    -1.41
    喜歡的
    -1.41
     своё
    -1.38
     ขาย
    -1.37
     acondicionado
    -1.34
     fár
    -1.34
    POSITIVE LOGITS
    𓏸
    1.95
    
    1.66
     ジュニア
    1.55
     />
    
    1.50
     シャンプー
    1.48
    horizontalLayout
    1.43
    _));
    1.43
     addirittura
    1.40
    </h2>
    1.39
    AutoModerator
    1.39
    Activations Density 0.299%

    No Known Activations