© 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. 80764
    Prev
    Next
    INDEX
    Explanations

    string concatenation and code

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

    This neuron responds to string‐building punctuation in code—specifically quotes and the concatenation operators (e.g. PHP’s “.” and SQL/JavaScript’s “+”).

    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
     from
    -1.25
     ostensibly
    -1.01
     vs
    -1.00
     What
    -0.99
    toDouble
    -0.98
     versus
    -0.90
     Makes
    -0.89
     sehr
    -0.89
    踵
    -0.88
     by
    -0.88
    POSITIVE LOGITS
     BRAD
    1.07
     konkret
    1.01
     MICHAEL
    1.01
     vay
    0.96
    特意
    0.96
     ALAN
    0.95
     mâle
    0.95
     insistence
    0.94
    ousing
    0.94
    的眼神
    0.93
    Activations Density 0.006%

    No Known Activations