© 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. 21189
    Prev
    Next
    INDEX
    Explanations

    closing punctuation or symbols

    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
    .setdefault
    -0.06
     Bread
    -0.06
    æıª
    -0.06
    lio
    -0.06
    çº
    -0.06
    gern
    -0.05
    åIJijè®°èĢħ
    -0.05
    éĻħ
    -0.05
    hoe
    -0.05
    aki
    -0.05
    POSITIVE LOGITS
    uld
    0.06
    æ½ľ
    0.06
    ntag
    0.05
    alc
    0.05
    puis
    0.05
    åįİ西
    0.05
    çİĭçĪ·
    0.05
     Dep
    0.05
    åĨħ容æľī
    0.05
    opf
    0.05
    Activations Density 0.004%

    No Known Activations