© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 39343
    Prev
    Next
    INDEX
    Explanations

    unexpected results or surprises

    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
    ä»®
    -0.08
    æĹłäºº
    -0.08
    èĩªåĬ©
    -0.08
    arka
    -0.08
     autof
    -0.08
     sep
    -0.08
    è¿ŁæĹ©
    -0.08
    é¸¡æ¯Ľ
    -0.08
    ecker
    -0.07
    åij¨ä¸Ģèĩ³
    -0.07
    POSITIVE LOGITS
     surprises
    0.12
     unexpected
    0.11
    ä¸įæĮī
    0.11
     surprising
    0.11
     unexpectedly
    0.11
     anom
    0.10
     surprise
    0.10
     anomal
    0.10
    åĩºä¹İ
    0.10
     sometimes
    0.10
    Activations Density 0.188%

    No Known Activations