© 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. 34-GEMMASCOPE-RES-131K
    4. 49995
    Prev
    Next
    INDEX
    Explanations

    replacing UI fragments

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

    This neuron strongly activates on programming-style code tokens (identifiers, keywords, and other code fragments) rather than ordinary prose.

    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
    Muss
    -0.75
    Fus
    -0.73
    yorsunuz
    -0.73
    Ralf
    -0.72
     autom
    -0.71
     decor
    -0.71
     Umwel
    -0.70
    Virgil
    -0.69
     annexed
    -0.69
    (',');
    -0.69
    POSITIVE LOGITS
    replace
    1.66
     replace
    1.56
    Fragment
    1.26
    Replace
    1.25
     replacement
    1.23
     replacing
    1.19
     replaced
    1.15
     frame
    1.14
     transaction
    1.13
    替换
    1.13
    Activations Density 0.013%

    No Known Activations