© 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. 28249
    Prev
    Next
    INDEX
    Explanations

    introduces specific items or definitions

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

    The neuron detects detailed electronics‐type technical references—especially mentions of components (like “diode”), circuit schematics, and numeric electrical measurements (voltage, current, inductance, coupling, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    ねー
    -1.11
     to
    -1.06
     or
    -1.05
     dėl
    -0.97
     like
    -0.92
     Us
    -0.92
     න
    -0.91
     gluing
    -0.91
    mon
    -0.90
    -
    -0.90
    POSITIVE LOGITS
     which
    1.44
    です
    1.30
    itisation
    1.19
     syk
    1.12
    CodeAttribute
    1.09
    gambarkan
    1.07
     yesterday
    1.05
    vesen
    1.05
     '\\;'
    1.05
     */
    1.04
    Activations Density 0.091%

    No Known Activations