© 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. 17377
    Prev
    Next
    INDEX
    Explanations

    numbers and counts

    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
    çł¥
    -0.06
    ún
    -0.06
    çŁ
    -0.06
    ania
    -0.05
    åĭī
    -0.05
     saf
    -0.05
    éĩıåĬĽ
    -0.05
    人æīĭ
    -0.05
    å°½éĩıå°ij
    -0.05
    mun
    -0.05
    POSITIVE LOGITS
    имоÑģÑĤÑĮ
    0.06
    ÙĤاش
    0.06
    URS
    0.06
     Coll
    0.05
    igur
    0.05
     coll
    0.05
    æĪijæĥ³
    0.05
    æ£
    0.05
    serter
    0.05
    imente
    0.05
    Activations Density 0.031%

    No Known Activations