© 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. 34-GEMMASCOPE-RES-131K
    7. 38059
    Prev
    Next
    INDEX
    Explanations

    foreign phrases and names

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

    This neuron fires on tokens that begin distinct sections or blocks—i.e. the first word of headings, code or license blocks, and document sections.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     frågor
    -0.89
    ㅡ
    -0.89
     lié
    -0.86
    ッチン
    -0.84
     dalších
    -0.84
    kkor
    -0.84
    Hm
    -0.83
    砷
    -0.83
    翳
    -0.83
    isée
    -0.82
    POSITIVE LOGITS
     people
    1.08
     two
    0.95
     presently
    0.91
     one
    0.91
    attled
    0.87
     current
    0.86
     Following
    0.85
     keep
    0.84
     errMsg
    0.84
    ądze
    0.83
    Activations Density 0.001%

    No Known Activations