© 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. 79913
    Prev
    Next
    INDEX
    Explanations

    initializing or empty structures

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     and
    -0.82
    tě
    -0.80
     étudié
    -0.79
     MyApp
    -0.75
     اگه
    -0.75
     högre
    -0.75
     дово
    -0.74
     increasingly
    -0.74
    ėjimo
    -0.74
    if
    -0.73
    POSITIVE LOGITS
     inicial
    1.33
     empty
    1.27
     initialize
    1.25
     initializes
    1.23
     default
    1.23
     initialise
    1.22
    初始
    1.21
     blank
    1.15
     boş
    1.15
     初始化
    1.13
    Activations Density 0.040%

    No Known Activations