© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 12513
    Prev
    Next
    INDEX
    Explanations

    resulting improvement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    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
    å®īåħ¨ç¬¬ä¸Ģ
    -0.07
    ä¹Łæĺ¯éľĢè¦ģ
    -0.07
    è¿ĺä¸įå¦Ĥ
    -0.07
    èĢĮä¸Ķè¦ģ
    -0.07
    以确å®ļ
    -0.07
    icz
    -0.07
    å¾Īä¸į
    -0.07
    æ¯Ķè¾ĥ严éĩį
    -0.07
    _intensity
    -0.07
     unless
    -0.06
    POSITIVE LOGITS
     improved
    0.13
    æĶ¹åĸĦäºĨ
    0.13
     improves
    0.13
    Improved
    0.12
    åĩıå°ijäºĨ
    0.12
     better
    0.11
    æıIJé«ĺäºĨ
    0.11
     improve
    0.11
     Improved
    0.11
     verbess
    0.11
    Activations Density 0.053%

    No Known Activations