© 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. 113899
    Prev
    Next
    INDEX
    Explanations

    code block start or inline notation

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

    LaTeX backslash commands, particularly at the start of command sequences.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on backslash tokens that introduce LaTeX commands.

    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
    
    -2.25
    
    -2.11
     própri
    -2.11
    蕖
    -2.08
     perioden
    -2.05
     parfü
    -2.02
    quelles
    -2.00
    鄚
    -2.00
    锨
    -2.00
     ANOTHER
    -1.99
    POSITIVE LOGITS
     used
    2.31
     was
    2.23
     regarded
    2.17
     When
    2.13
     in
    2.06
     only
    2.03
     asked
    2.00
     made
    1.99
     had
    1.98
     turned
    1.96
    Activations Density 0.004%

    No Known Activations