© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21124
    Prev
    Next
    INDEX
    Explanations

    special characters followed by questions

    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
    æĺ¯èĢģ
    -0.05
    çĥŁ
    -0.05
     assured
    -0.05
    çIJĨ
    -0.05
    usic
    -0.05
    run
    -0.04
     signed
    -0.04
    ilt
    -0.04
    ÏģιÏĥ
    -0.04
     fancy
    -0.04
    POSITIVE LOGITS
     пиÑĤом
    0.06
    ãģijãģ©ãģŃ
    0.06
    kov
    0.06
     диÑģкомÑĦоÑĢ
    0.06
    èĢIJåıĹ
    0.06
    éĢļè¿ĩ使çĶ¨
    0.06
    nyt
    0.06
    Cou
    0.06
    isting
    0.06
    oust
    0.06
    Activations Density 0.205%

    No Known Activations