© 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. 21961
    Prev
    Next
    INDEX
    Explanations

    marker tokens like [@B, [@CR

    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
    	   ↵
    -0.06
    æµģè¿ŀ
    -0.05
      ↵
    -0.05
    éŸ
    -0.05
     afinal
    -0.05
    atson
    -0.05
    æ½Ķ
    -0.05
    å®ŀä½ĵ
    -0.05
    åĬ¨ä¸įåĬ¨
    -0.05
    ockey
    -0.05
    POSITIVE LOGITS
    0.07
     determination
    0.06
    chts
    0.06
    ç¿
    0.06
    ä¸Ĭå½ĵ
    0.05
    ajang
    0.05
    -mails
    0.05
    nota
    0.05
    luca
    0.05
     insol
    0.05
    Activations Density 0.010%

    No Known Activations