© 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. 22-GEMMASCOPE-RES-131K
    7. 130540
    Prev
    Next
    INDEX
    Explanations

    The neuron detects technical specification terms and quantifiers—words that convey numeric parameters or structured measurements in engineering‐style descriptions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    describes technical connections and parts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     myriad
    -0.98
     Semoga
    -0.90
    строить
    -0.89
     mojej
    -0.88
     lauded
    -0.87
     этих
    -0.86
     extraño
    -0.85
     charité
    -0.85
    giarism
    -0.84
     insanely
    -0.84
    POSITIVE LOGITS
     November
    0.94
     still
    0.94
     "/";
    0.92
    ♭
    0.89
     │
    0.88
     ensi
    0.87
     sala
    0.87
     ajustable
    0.86
     again
    0.86
     December
    0.86
    Activations Density 0.097%

    No Known Activations