© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 40093
    Prev
    Next
    INDEX
    Explanations

    tobacco and cigarettes

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

    tobacco and smoking related terminology.

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

    The neuron consistently activates on mentions of tobacco use and smoking (e.g. “tobacco,” “smoking,” “smoke”).

    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
     
    -1.63
    の説明
    -1.19
    {
    -1.17
     /
    -1.16
    動か
    -1.16
    li
    -1.14
     \
    -1.14
     of
    -1.13
    ↵
    -1.12
     helfen
    -1.10
    POSITIVE LOGITS
     cigarettes
    1.59
     aluminio
    1.38
     tobacco
    1.36
     rokok
    1.34
    Hvordan
    1.34
    itinéraire
    1.31
     cigarette
    1.30
    smoking
    1.30
    猡
    1.27
    haling
    1.27
    Activations Density 0.044%

    No Known Activations