© 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. 61386
    Prev
    Next
    INDEX
    Explanations

    non-latin or punctuation followed by phrase start

    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
    perf
    -0.07
    ien
    -0.07
    åģıåģı
    -0.06
    nat
    -0.06
    ork
    -0.06
    lic
    -0.06
    agi
    -0.06
    lining
    -0.06
    icul
    -0.06
    æľªçŁ¥
    -0.06
    POSITIVE LOGITS
    è¿Ļ个åĽŀçŃĶ
    0.09
    uhe
    0.08
    .proxy
    0.07
    Ñĸз
    0.07
     unnecessarily
    0.07
    è¿Ļæł·è¯´
    0.07
    evol
    0.07
    ventus
    0.07
    oshi
    0.07
    ejo
    0.07
    Activations Density 0.009%

    No Known Activations