© 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. 11879
    Prev
    Next
    INDEX
    Explanations

    internal mechanisms and electronics

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

    The neuron fires on dense, technical descriptions of precision instrument components—e.g. detailed terminology like “sensitive element,” “gyro,” “rotor,” and related parenthetical definitions.

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

    The neuron fires strongly on the specialized technical‐specification vocabulary and formatting (e.g. component names, parenthetical “e.g.” clauses, “includes,” “the,” etc.) characteristic of patent‐style descriptive passages.

    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
     vícti
    -0.98
    kleid
    -0.88
    strator
    -0.86
    shutterstock
    -0.83
    ὺς
    -0.81
    laptop
    -0.81
    tský
    -0.81
    zeugen
    -0.81
    ésie
    -0.80
    етка
    -0.80
    POSITIVE LOGITS
     internal
    2.25
    内部
    2.14
     internals
    1.98
    internal
    1.87
     mechanism
    1.81
     mechanisms
    1.57
    Internal
    1.56
     circuitry
    1.55
     internally
    1.47
     electronics
    1.46
    Activations Density 0.153%

    No Known Activations