© 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. 34-GEMMASCOPE-RES-131K
    4. 82260
    Prev
    Next
    INDEX
    Explanations

    company names and organizations

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

    The neuron detects the initial subword segments of proper names—especially the leading parts of company, product, or organization names.

    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
    Máy
    -0.89
     tunik
    -0.88
    =“
    -0.84
    鍾
    -0.84
    ▶
    -0.81
     hypnotic
    -0.79
     clf
    -0.79
    mydb
    -0.78
    corollary
    -0.77
     dör
    -0.77
    POSITIVE LOGITS
    color
    1.01
    mostly
    0.96
    col
    0.95
    whose
    0.95
    can
    0.90
    no
    0.90
    tex
    0.90
    parse
    0.90
    ac
    0.89
    serve
    0.88
    Activations Density 0.070%

    No Known Activations