© 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. 17742
    Prev
    Next
    INDEX
    Explanations

    rewritten, open

    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
    azen
    -0.06
    æ¡ij
    -0.05
    åľ°çIJĥä¸Ĭ
    -0.05
    asse
    -0.05
    éĻĦçĿĢ
    -0.05
    atte
    -0.05
    åħī
    -0.05
    Ø·
    -0.05
    usto
    -0.05
     taps
    -0.05
    POSITIVE LOGITS
    ergestellt
    0.07
    timi
    0.06
    конÑģÑĥлÑĮÑĤи
    0.06
    еÑģÑĮ
    0.06
    мали
    0.06
    éĤ£åIJį
    0.06
     ÐłÑĥÑģ
    0.06
     Croat
    0.06
     (?)
    0.06
    åħ¬åı·
    0.06
    Activations Density 0.003%

    No Known Activations