© 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. 20370
    Prev
    Next
    INDEX
    Explanations

    question marks and periods

    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
     noix
    -0.06
     nových
    -0.06
    åĬ²
    -0.06
    åĨ³å®ļæĺ¯åIJ¦
    -0.06
     alegr
    -0.06
    ÑĢоÑģло
    -0.06
    anh
    -0.06
     prud
    -0.06
    lot
    -0.05
    oug
    -0.05
    POSITIVE LOGITS
    TEE
    0.06
    izen
    0.05
    IFE
    0.05
    asm
    0.05
     legit
    0.05
    æİ¢
    0.05
    æŁ¥
    0.05
    lore
    0.05
     definit
    0.04
     bed
    0.04
    Activations Density 0.003%

    No Known Activations