© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 15981
    Prev
    Next
    INDEX
    Explanations

    death and finality

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

    The neuron detects words and phrases related to death or dying and end‐of‐life contexts.

    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
    總是
    -1.06
    纔
    -1.05
     ruines
    -1.01
    rist
    -0.97
     Werden
    -0.96
     Agamemnon
    -0.94
     vectorstock
    -0.93
    んだから
    -0.93
     ны
    -0.92
     Erste
    -0.92
    POSITIVE LOGITS
     dying
    1.61
     अं
    1.34
    最后
    1.31
     before
    1.25
     terakhir
    1.23
    最後
    1.22
     death
    1.22
    ització
    1.20
     final
    1.20
    żdy
    1.19
    Activations Density 0.022%

    No Known Activations