© 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. 63701
    Prev
    Next
    INDEX
    Explanations

    whether correct

    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
    åĪļä¸Ģ
    -0.08
    emani
    -0.07
    ä¹Łä¸įå¤ļ
    -0.07
    ÏĦζ
    -0.07
     alun
    -0.07
    éĩįè¦ģçļĦäºĭæĥħ
    -0.07
    stown
    -0.07
    主åĭķ
    -0.07
    æĥĬ
    -0.07
    IMPORTANT
    -0.07
    POSITIVE LOGITS
    æĺ¯åIJ¦ç¬¦åIJĪ
    0.11
    æĺ¯åIJ¦æŃ£ç¡®
    0.11
     correctness
    0.11
    对éĶĻ
    0.11
     validity
    0.10
     whether
    0.10
    æĺ¯åIJ¦
    0.09
    æĺ¯åIJ¦ä¸º
    0.09
    羣åģĩ
    0.09
    åĩĨç¡®æĢ§
    0.08
    Activations Density 0.031%

    No Known Activations