© 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. 10-GEMMASCOPE-RES-131K
    4. 105641
    Prev
    Next
    INDEX
    Explanations

    programming code and syntax

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

    function or method signatures in programming code, particularly Swift and Objective-C method declarations with parameters.

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

    This neuron responds to Swift function signatures, especially the return arrow (`->`) and the declared return type (`Int`), as well as related tokens in table-view method declarations.

    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
     Their
    -1.61
    Their
    -1.54
     compreender
    -1.52
     quaisquer
    -1.40
     their
    -1.40
    什么事
    -1.37
    该如何
    -1.37
     can
    -1.35
     conhecer
    -1.29
     disposiciones
    -1.29
    POSITIVE LOGITS
            
    1.44
                   
    1.33
           
    1.27
                     
    1.25
    ynchronously
    1.24
                       
    1.23
     verduras
    1.23
               
    1.22
    harapkan
    1.20
                 
    1.17
    Activations Density 0.015%

    No Known Activations