© 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. 19802
    Prev
    Next
    INDEX
    Explanations

    study

    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
    ÙĨدÙĩ
    -0.07
    ed
    -0.06
    ei
    -0.06
    ers
    -0.06
    h
    -0.06
    bey
    -0.05
    o
    -0.05
    care
    -0.05
     szere
    -0.05
    en
    -0.05
    POSITIVE LOGITS
    íĻĶ
    0.06
    iony
    0.06
    ippy
    0.06
    åĽŃè·¯
    0.06
    posa
    0.06
    azia
    0.06
    با
    0.06
     Archived
    0.06
    åĮĸåĴĮ
    0.06
     "\",
    0.05
    Activations Density 0.001%

    No Known Activations