© 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. 112771
    Prev
    Next
    INDEX
    Explanations

    sleep disruption or apnea

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

    words and phrases related to sleep and sleep-related conditions.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates strongly on the word “sleep” (and its variants like “sleeping”), identifying mentions of sleep-related terms.

    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
    籏
    -2.59
    翃
    -2.56
     применя
    -2.55
    ブラシ
    -2.53
    肏
    -2.48
    会出现
    -2.47
     integridad
    -2.39
    僦
    -2.34
    致力
    -2.34
    が付いて
    -2.33
    POSITIVE LOGITS
    -
    3.81
    或者
    3.08
     یا
    3.02
    驁
    2.42
     или
    2.39
     In
    2.39
     weitgehend
    2.33
     atau
    2.27
    there
    2.20
    pulseira
    2.20
    Activations Density 0.007%

    No Known Activations