© 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. 18608
    Prev
    Next
    INDEX
    Explanations

    numbers following 's

    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
    led
    -0.05
     isa
    -0.05
    ollo
    -0.05
    sy
    -0.05
    akt
    -0.05
    ÙĬÙĨا
    -0.05
    ste
    -0.05
    æŀļ
    -0.05
    so
    -0.05
    —we
    -0.05
    POSITIVE LOGITS
     ìŀĩ
    0.06
    egend
    0.06
    缮ä¸Ń
    0.05
    溪éķĩ
    0.05
    åĪ¥
    0.05
    ekh
    0.05
    _registers
    0.05
     mue
    0.05
    æĶ¯æĮģä¸ĭ
    0.05
    大æĪ·
    0.05
    Activations Density 0.224%

    No Known Activations