© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 64306
    Prev
    Next
    INDEX
    Explanations

    and followed by a verb

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

    This neuron responds to discourse-level connective and emphasis words—especially conjunctions and adverbs like “and,” “particularly,” “especially,” “so,” and similar markers that introduce or emphasize new information.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    licitations
    -1.31
    PENDAHULUAN
    -1.30
    תודה
    -1.30
    χεία
    -1.30
    ₈
    -1.27
     suéter
    -1.24
    ()]
    -1.20
     efekty
    -1.20
    ומית
    -1.20
    ต่างๆ
    -1.19
    POSITIVE LOGITS
              
    1.73
          
    1.73
        
    1.69
           
    1.68
                                   
    1.67
         
    1.64
     (
    1.58
    =
    1.57
      
    1.55
       
    1.51
    Activations Density 0.054%

    No Known Activations