© 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. 28202
    Prev
    Next
    INDEX
    Explanations

    returning code values

    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
     retourner
    -1.08
    ##
    -1.05
    fucking
    -0.98
    BETWEEN
    -0.96
     fidé
    -0.95
     GameManager
    -0.95
    点了点头
    -0.94
    腠
    -0.94
    XmlAccessorType
    -0.93
    zf
    -0.93
    POSITIVE LOGITS
     first
    1.01
     or
    1.00
    亮点
    0.98
    ตรี
    0.94
     allows
    0.92
    还得
    0.91
     serves
    0.90
     enables
    0.89
     helps
    0.89
     cheerful
    0.88
    Activations Density 0.001%

    No Known Activations