© 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. 18647
    Prev
    Next
    INDEX
    Explanations

    instead or rather

    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
    ickers
    -0.08
    缸åħ³æݨèįIJ
    -0.08
    æĥ³å¾Ĺ
    -0.07
    çī¹åľ°
    -0.07
    ablish
    -0.07
    áĥŃ
    -0.07
    è²Įä¼¼
    -0.07
    ä¸įåı¯èĥ½çļĦ
    -0.07
    aint
    -0.07
    ammt
    -0.07
    POSITIVE LOGITS
     instead
    0.17
     rather
    0.16
    èĢĮä¸įæĺ¯
    0.16
    instead
    0.16
    èĢĮéĿŀ
    0.14
    Instead
    0.14
     Instead
    0.13
    rather
    0.13
     plutôt
    0.13
     Rather
    0.13
    Activations Density 0.545%

    No Known Activations