© 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. 34-GEMMASCOPE-RES-131K
    7. 50474
    Prev
    Next
    INDEX
    Explanations

    alternative definitions

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

    This neuron fires on code‐like tokens—especially identifiers, type names, macros, and symbols (often containing uppercase letters, underscores, or punctuation) in programming excerpts.

    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
    ṙ
    -0.95
    嚐
    -0.90
     modus
    -0.88
    flüs
    -0.88
     レスト
    -0.88
     mirador
    -0.88
     personel
    -0.86
     brillant
    -0.86
     existant
    -0.85
    Ḩ
    -0.85
    POSITIVE LOGITS
     another
    1.53
     really
    1.16
     bigger
    1.11
     secondary
    1.08
     second
    1.03
    another
    1.01
     mini
    1.00
     alt
    1.00
     _
    0.99
     also
    0.97
    Activations Density 0.096%

    No Known Activations