© 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. 8876
    Prev
    Next
    INDEX
    Explanations

    exhausted or exhaustive use

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

    The neuron activates on words derived from the root “exhaust” (e.g. exhaust, exhausted, exhaustive, exhaustion, or the phrase “exhaust administrative remedies”).

    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
     ACKNOWLEDGMENTS
    -2.17
    銩
    -2.17
    }$
    -2.16
     ´
    -2.03
     ʻ
    -2.02
     maven
    -2.02
    」
    -1.98
      
    -1.95
    釹
    -1.95
    seems
    -1.92
    POSITIVE LOGITS
     but
    2.31
     elaborate
    2.02
    ～～
    1.95
     invigor
    1.94
     ONLY
    1.94
    浓郁
    1.90
    极致
    1.90
     exuberant
    1.88
     rapport
    1.86
    想要的
    1.81
    Activations Density 0.007%

    No Known Activations