© 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. 22-GEMMASCOPE-RES-131K
    7. 65583
    Prev
    Next
    INDEX
    Explanations

    deterioration or defects over time

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

    The neuron detects language describing gradual degradation or wearing out (damage, deterioration, or aging) of something over time.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     conhece
    -1.17
    無駄
    -1.16
    みたら
    -1.12
     épaules
    -1.09
     tivesse
    -1.08
    ımda
    -1.05
    寝る
    -1.05
    ছে
    -1.02
     ujarnya
    -1.00
     folos
    -0.97
    POSITIVE LOGITS
     becomes
    1.59
     irgendwann
    1.47
     gradually
    1.38
     become
    1.38
     Eventually
    1.27
     تد
    1.25
     gets
    1.09
    Eventually
    1.09
     will
    1.08
     after
    1.08
    Activations Density 0.105%

    No Known Activations