© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 68371
    Prev
    Next
    INDEX
    Explanations

    Terminal, Culture, jeff, Category

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

    This neuron detects hyperlink-related tokens—especially anchor text like “official,” “website,” “learn more,” and domain/URL fragments.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     maleta
    -1.09
    
    -1.05
     intimid
    -1.05
    ኧ
    -1.05
     пищи
    -1.03
     Galería
    -1.02
     baterie
    -1.02
     desn
    -0.96
     haue
    -0.96
     mês
    -0.95
    POSITIVE LOGITS
     its
    1.02
    imakasih
    0.99
     of
    0.98
     Orte
    0.96
     on
    0.96
     coincides
    0.86
     вось
    0.84
    之旅
    0.84
    IVID
    0.82
    झे
    0.82
    Activations Density 0.012%

    No Known Activations