© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 12878
    Prev
    Next
    INDEX
    Explanations

    even + verb

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

    The neuron responds to very common high-frequency English words—especially short function or auxiliary words like “has,” “a,” “been,” “led,” and “used.”

    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
    \
    -2.50
     
    -2.08
      
    -2.05
          
    -1.85
    also
    -1.79
     \
    -1.78
    ”;
    -1.77
                
    -1.75
         
    -1.75
       
    -1.72
    POSITIVE LOGITS
     esfuer
    2.09
    ൅
    1.91
    秫
    1.89
    𖤣
    1.88
    ਲ਼
    1.81
    是怎么回事
    1.76
    眎
    1.73
    是怎么
    1.73
     possiamo
    1.71
     convocatoria
    1.70
    Activations Density 0.014%

    No Known Activations