© 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. 18962
    Prev
    Next
    INDEX
    Explanations

    Advanced, AE, Classic, Plug, Bus, Plus

    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
    ][(
    -0.06
     ness
    -0.06
    irt
    -0.05
    ear
    -0.05
    YTE
    -0.05
    ersten
    -0.05
    IRT
    -0.05
    èįĴ
    -0.05
    èĢģå®ŀ
    -0.05
    cade
    -0.05
    POSITIVE LOGITS
    alah
    0.06
    repr
    0.06
    åįĥåı¤
    0.06
    æľ¬æĬ¥
    0.06
    wechsel
    0.05
    outines
    0.05
     disob
    0.05
    ÑĤана
    0.05
    Ø·Ùģ
    0.05
    è°ĪæģĭçĪ±
    0.05
    Activations Density 0.022%

    No Known Activations