© 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. 21346
    Prev
    Next
    INDEX
    Explanations

    capitals and countries

    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
     ÑģовмеÑģÑĤи
    -0.08
     bada
    -0.08
     prévo
    -0.08
     aktivit
    -0.08
    ultat
    -0.08
     partn
    -0.08
     Kontaktd
    -0.08
     алÑĮÑĤеÑĢна
    -0.07
     rentrée
    -0.07
     descubr
    -0.07
    POSITIVE LOGITS
    0.07
     non
    0.05
     a
    0.05
     ch
    0.05
    !
    0.05
     e
    0.05
     '
    0.05
     u
    0.05
     min
    0.05
     tr
    0.05
    Activations Density 0.012%

    No Known Activations