© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 20095
    Prev
    Next
    INDEX
    Explanations

    drop or dropped

    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
    usion
    -0.06
    측
    -0.06
    樽
    -0.05
    å¤ļæĸ¹
    -0.05
    à¸ķรà¸ĩ
    -0.05
    urement
    -0.05
    irement
    -0.05
    ago
    -0.05
    -0.05
    æĸ¹çļĦ
    -0.05
    POSITIVE LOGITS
    lish
    0.07
     thừa
    0.07
    spot
    0.06
     кÑĥлÑĮ
    0.06
    ÑĪÑĤа
    0.06
    ждÑĥÑİ
    0.06
     spot
    0.06
    sville
    0.06
    ongan
    0.06
    èIJ½èĦļ
    0.06
    Activations Density 0.008%

    No Known Activations