© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 8727
    Prev
    Next
    INDEX
    Explanations

    DOE, urlencoded, intelligence, education

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

    This neuron fires most strongly on occurrences of “DOE” (the U.S. Department of Energy) and closely related energy-agency mentions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    哢
    -2.67
    ?”
    -2.59
    if
    -2.52
    ”;
    -2.30
    ”,
    -2.30
    )
    -2.27
    ).”
    -2.25
    !”
    -2.16
     '*':
    -2.05
    閦
    -2.05
    POSITIVE LOGITS
     of
    3.13
     tetapi
    2.25
    я
    2.17
     tujuh
    2.02
    းက
    1.98
     We
    1.97
    および
    1.94
    (
    1.88
    >
    1.85
    趼
    1.80
    Activations Density 0.002%

    No Known Activations