© 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. 73465
    Prev
    Next
    INDEX
    Explanations

    Inner Mongolia

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

    The neuron is a high-frequency token detector—it lights up strongest on very common English function words (e.g. “the,” “on,” “a,” 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_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
     Malaysian
    -0.95
     duck
    -0.81
     bamboo
    -0.78
    arsch
    -0.76
     Taiwanese
    -0.75
     عل
    -0.75
     creeks
    -0.75
    镌
    -0.75
     elective
    -0.75
     instrucción
    -0.74
    POSITIVE LOGITS
     Desert
    0.99
    Wall
    0.96
     Wall
    0.92
    Desert
    0.88
    wall
    0.88
     Great
    0.86
    Northern
    0.86
     deserts
    0.82
     northern
    0.81
    Inner
    0.80
    Activations Density 0.026%

    No Known Activations