© 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. 21477
    Prev
    Next
    INDEX
    Explanations

    closing punctuation or braces

    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
    erken
    -0.07
    æĥħçļĦ
    -0.07
    å°º
    -0.07
     Näh
    -0.06
     ÙħتÙħ
    -0.06
    æĥħ
    -0.06
    綴
    -0.06
    측
    -0.06
    หัà¸ģ
    -0.06
    æķ´ä¸ªéĹ®é¢ĺ
    -0.06
    POSITIVE LOGITS
    one
    0.06
    res
    0.05
    ray
    0.05
    ck
    0.05
    f
    0.05
    run
    0.05
    CL
    0.05
    ONE
    0.05
    Ch
    0.05
    çĤ¹å¯¹
    0.05
    Activations Density 0.032%

    No Known Activations