© 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. 47909
    Prev
    Next
    INDEX
    Explanations

    middleware component descriptions

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

    This neuron detects long, all-caps identifiers with underscore separators, such as C/C++ preprocessor guard macros.

    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
     Leighton
    -0.79
    Allister
    -0.71
    탄
    -0.70
     ติด
    -0.69
     gorilla
    -0.67
    romat
    -0.67
    TokenNameLBRACE
    -0.66
     scelto
    -0.65
     condominium
    -0.65
     elbow
    -0.64
    POSITIVE LOGITS
     middleware
    1.13
    IDL
    1.13
     IDL
    1.06
    idl
    1.01
     COR
    0.99
    omg
    0.97
     Middleware
    0.95
    ORB
    0.93
     OMG
    0.89
     DDS
    0.84
    Activations Density 0.041%

    No Known Activations