© 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. 34-GEMMASCOPE-RES-131K
    7. 116247
    Prev
    Next
    INDEX
    Explanations

    legal documents and code structure

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

    The neuron fires on specialized or domain‐specific tokens—uncommon proper names, technical identifiers (like class or interface names), CSS properties, legal citation numbers, and other rare jargon.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     budou
    -0.73
    Receive
    -0.72
    icom
    -0.71
     emberek
    -0.68
     democrati
    -0.68
    Ler
    -0.68
    ண்ண
    -0.68
     submer
    -0.68
     Bobby
    -0.66
     Peacock
    -0.66
    POSITIVE LOGITS
    义
    0.73
    龥
    0.69
    これまで
    0.68
     Dishwasher
    0.68
     Cervantes
    0.67
     ruolo
    0.67
     سرخ
    0.67
    ೋ
    0.67
     Etr
    0.66
     archivio
    0.65
    Activations Density 0.097%

    No Known Activations