© 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. 17234
    Prev
    Next
    INDEX
    Explanations

    who-know placeholders from above

    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
    rief
    -0.05
    uyer
    -0.05
    é¦Ī
    -0.05
    arat
    -0.05
    ervis
    -0.05
     تض
    -0.05
    æŃ¢
    -0.04
    æ·»
    -0.04
     Bueno
    -0.04
    æľī幸
    -0.04
    POSITIVE LOGITS
    æľīå¿ĥ
    0.07
    çİĭä¹ĭ
    0.06
    ÑĤеÑĢÑĥ
    0.06
    æĢİ
    0.06
    模å¼ıä¸Ń
    0.06
    ]=$
    0.06
    жан
    0.06
    çĶ¨ä»Ģä¹Ī
    0.06
    åīį款
    0.06
    zia
    0.06
    Activations Density 0.001%

    No Known Activations