© 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. 22-GEMMASCOPE-RES-131K
    7. 28537
    Prev
    Next
    INDEX
    Explanations

    ViewController

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

    This neuron fires on capitalized programming identifiers—especially framework/module names, header filenames, and class/type names in imports and interfaces.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     дъ
    -1.00
     delikat
    -0.98
    philly
    -0.97
    交往
    -0.96
     gekenn
    -0.96
     trabajando
    -0.96
    owm
    -0.95
     sufre
    -0.95
     mantend
    -0.94
    thanks
    -0.94
    POSITIVE LOGITS
     tost
    1.09
     adorno
    1.05
    本书
    1.04
    Михаил
    1.04
     soleil
    1.02
     cabine
    0.99
     carnaval
    0.98
    咻
    0.98
     silueta
    0.97
    说起
    0.97
    Activations Density 0.019%

    No Known Activations