© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 18881
    Prev
    Next
    INDEX
    Explanations

    but, however, though, except

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     دع
    -0.05
    æĹłè¾ľ
    -0.05
    bod
    -0.05
     âŁ
    -0.05
     Write
    -0.05
    ä½ĵåĴĮ
    -0.05
     dun
    -0.05
    æŁIJ个
    -0.05
     bodies
    -0.05
     somewhere
    -0.05
    POSITIVE LOGITS
    ost
    0.06
    ict
    0.06
    ostal
    0.06
    awan
    0.06
     cấp
    0.06
    anzi
    0.05
    umbing
    0.05
    è¿Ľè¡ĮæŶ
    0.05
     run
    0.05
    ään
    0.05
    Activations Density 0.042%

    No Known Activations