© 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. 21520
    Prev
    Next
    INDEX
    Explanations

    version information

    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
    åħ¶ä»ĸ人çļĦ
    -0.07
    åĴĮåħ¶ä»ĸ
    -0.06
     saja
    -0.05
    hrad
    -0.05
    åıĬåħ¶ä»ĸ
    -0.05
    kových
    -0.05
     Fá
    -0.05
    ä¸įåĮħæĭ¬
    -0.05
    åħ¶ä»ĸåªĴä½ĵ
    -0.05
    alama
    -0.05
    POSITIVE LOGITS
    tracts
    0.06
    ç»ıçĶ±
    0.06
    izedName
    0.06
    éĢļè´§
    0.05
     consistente
    0.05
    nal
    0.05
    edList
    0.05
     urat
    0.05
    tracted
    0.05
    çļĦ信念
    0.05
    Activations Density 0.001%

    No Known Activations