© 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. 20421
    Prev
    Next
    INDEX
    Explanations

    I know what

    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
    er
    -0.06
    æĶ
    -0.06
     Likely
    -0.05
     likely
    -0.05
     fixture
    -0.05
    aka
    -0.05
     specifying
    -0.05
     Äijạo
    -0.05
    大éĩıçļĦ
    -0.05
     LEFT
    -0.05
    POSITIVE LOGITS
    oogle
    0.07
    CardContent
    0.06
    OSA
    0.06
    çıĪ
    0.06
    emmän
    0.06
    )./
    0.06
    iaal
    0.06
    endlich
    0.06
    ä¸įéĢı
    0.06
    azzo
    0.06
    Activations Density 0.002%

    No Known Activations