© 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. 65547
    Prev
    Next
    INDEX
    Explanations

    lifestyle and behavior factors

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

    The neuron detects mentions of modifiable risk factors or lifestyle/behavioral determinants (e.g., “modifiable,” “lifestyle,” “behavior”).

    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
     UIFont
    -1.15
     沖縄
    -1.12
    いため
    -1.05
    wość
    -1.03
    接続
    -1.03
    juje
    -1.01
     дѣ
    -1.01
     różne
    -0.98
    Kegiatan
    -0.97
     heerlijk
    -0.96
    POSITIVE LOGITS
     lifestyle
    1.98
    IFESTYLE
    1.55
     diet
    1.41
     recommendations
    1.36
     desmotiv
    1.34
     recommended
    1.29
     choices
    1.28
     AppColors
    1.28
     habits
    1.26
     Lifestyle
    1.26
    Activations Density 0.084%

    No Known Activations