© 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. 73390
    Prev
    Next
    INDEX
    Explanations

    different programming language constructs

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

    The neuron is activating on programming‐or configuration‐style tokens—file names, identifiers, attribute names, version numbers, CSS/HTML references, and other code/configuration keywords.

    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
     Beams
    -0.73
    Share
    -0.71
     sharing
    -0.69
    entee
    -0.68
     ammat
    -0.68
    Sucesor
    -0.67
    Lead
    -0.67
    któ
    -0.66
    nuts
    -0.66
     FEBRUARY
    -0.65
    POSITIVE LOGITS
    anso
    0.72
    Personajes
    0.66
    Shops
    0.65
    chein
    0.64
    Slightly
    0.64
    Artículos
    0.64
     люб
    0.62
     TBD
    0.61
     prévention
    0.61
    Terri
    0.61
    Activations Density 0.081%

    No Known Activations