© 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. 18259
    Prev
    Next
    INDEX
    Explanations

    surprise gifts, visits, or news

    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
    uncate
    -0.06
    ibera
    -0.06
    éĶĭåĪ©çļĦ
    -0.06
    ipper
    -0.05
    æļ¢
    -0.05
    selling
    -0.05
    ême
    -0.05
    ÙĪاÙĨÙĬ
    -0.05
    çļĦåĪ©çĽĬ
    -0.05
    lost
    -0.05
    POSITIVE LOGITS
     surprises
    0.07
     surprise
    0.06
    çªģçĦ¶
    0.06
    æĥĬåĸľ
    0.06
    ä¸įçŁ¥æīĢ
    0.06
    忽çĦ¶
    0.06
     Surprise
    0.06
    æį§
    0.05
    åľ°å¹³
    0.05
     заÑĢанее
    0.05
    Activations Density 0.035%

    No Known Activations