© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 71300
    Prev
    Next
    INDEX
    Explanations

    specific term contexts

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

    The neuron responds most strongly to unusual or low-frequency tokens—long, specialized words (and even rare punctuation sequences like “…”) that stand out in a document.

    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
    страции
    -1.04
    Люди
    -1.02
    بيس
    -1.02
     גרם
    -1.01
    劏
    -1.00
     ILLUSTRATIONS
    -1.00
    哟
    -0.99
     AGRICULTURE
    -0.98
    ového
    -0.98
    SHIRE
    -0.97
    POSITIVE LOGITS
     Änder
    1.07
    我们
    1.05
    it
    0.96
    我们就
    0.96
     Our
    0.96
    „
    0.96
     entiende
    0.95
     l
    0.94
     PR
    0.93
     „
    0.93
    Activations Density 0.000%

    No Known Activations