© 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. 27509
    Prev
    Next
    INDEX
    Explanations

    rather or implying

    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
    _Call
    -0.08
    æŃ¤èµ·
    -0.08
    åĬłåĭĴ
    -0.07
    lott
    -0.07
    edir
    -0.07
    ocator
    -0.07
    åģļçļĦå°±æĺ¯
    -0.07
    ÙĤÙħ
    -0.07
    åĨ·åĨ·çļĦ
    -0.07
    culares
    -0.07
    POSITIVE LOGITS
    çļĦæĦıæĢĿ
    0.12
     rather
    0.11
    rather
    0.10
    ãģ¨ãģĦãģĨæĦıåij³
    0.10
    èĢĮä¸įæĺ¯
    0.10
     Rather
    0.09
    æĦıæĢĿæĺ¯
    0.09
    æĦıæĢĿ
    0.09
     implying
    0.08
     plutôt
    0.08
    Activations Density 0.021%

    No Known Activations