© 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. 10-GEMMASCOPE-RES-131K
    7. 68398
    Prev
    Next
    INDEX
    Explanations

    proper nouns and specific terms

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

    The neuron fires on uncommon or rare word‐pieces—especially longer, multi-syllable or capitalized tokens you typically see in titles, headings, technical terms, or proper names.

    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
     would
    -2.31
     such
    -1.67
     new
    -1.63
    Since
    -1.53
     like
    -1.52
    Following
    -1.49
    <td>
    -1.48
    ↵↵↵
    -1.48
     while
    -1.45
    They
    -1.45
    POSITIVE LOGITS
     durabilidad
    1.68
     calentamiento
    1.55
     then
    1.51
     göm
    1.49
    
    1.48
    分に
    1.48
    很赞
    1.48
    っていると
    1.48
     propietarios
    1.45
     estacionamiento
    1.45
    Activations Density 0.125%

    No Known Activations