© 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. 39718
    Prev
    Next
    INDEX
    Explanations

    code assignments and declarations

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

    The neuron flags occurrences of user‐defined identifiers (variable, constant, or other named symbols) in code snippets.

    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
     empezar
    -1.06
     comenzar
    -1.05
    Somehow
    -1.02
     esquecer
    -0.97
     colgar
    -0.96
     aguj
    -0.96
    ınızı
    -0.95
     reklama
    -0.95
     zawsze
    -0.94
     kwiet
    -0.94
    POSITIVE LOGITS
     just
    0.99
     not
    0.98
     LSM
    0.96
     גם
    0.95
     Tern
    0.94
     crux
    0.92
     Tracker
    0.90
     SGS
    0.90
     sowohl
    0.89
    ्यान
    0.89
    Activations Density 0.001%

    No Known Activations