© 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. 22-GEMMASCOPE-RES-131K
    4. 68464
    Prev
    Next
    INDEX
    Explanations

    code snippets

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

    The neuron fires on identifier tokens—especially short variable or parameter names—within code.

    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
     not
    -0.99
     Often
    -0.96
     while
    -0.96
    Kedua
    -0.90
    träd
    -0.90
    Ikus
    -0.88
     epitom
    -0.87
     on
    -0.86
    ekš
    -0.86
     delivers
    -0.85
    POSITIVE LOGITS
     padrões
    1.05
     incluindo
    1.03
    だら
    0.97
     caráter
    0.97
     doenças
    0.95
     créer
    0.94
     estoque
    0.93
    ulic
    0.93
     woda
    0.93
     colorir
    0.93
    Activations Density 0.027%

    No Known Activations