© 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. 20160
    Prev
    Next
    INDEX
    Explanations

    division and categorization

    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
    oral
    -0.07
    à¸ģารà¸ĵà¹Į
    -0.06
    ä¸Ģ个
    -0.06
    TEL
    -0.06
    ponent
    -0.06
    ä¸Ģ代
    -0.06
    ä¸Ģå¥Ĺ
    -0.06
    oj
    -0.06
    usunan
    -0.05
    ijk
    -0.05
    POSITIVE LOGITS
    ality
    0.06
     بÙĨدÛĮ
    0.06
    еп
    0.06
    s
    0.06
    schuh
    0.06
    ided
    0.05
     Balik
    0.05
    yard
    0.05
    çļĦçĥŃæĥħ
    0.05
    подоб
    0.05
    Activations Density 0.036%

    No Known Activations