© 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. 18138
    Prev
    Next
    INDEX
    Explanations

    if condition

    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
    ģ
    -0.06
    ime
    -0.06
     بدÙĦ
    -0.06
    elis
    -0.05
    å¯Ĩä¸įåı¯åĪĨ
    -0.05
    تÙĪÙĨ
    -0.05
    ếm
    -0.05
    -sizing
    -0.05
    aque
    -0.05
    èij«
    -0.05
    POSITIVE LOGITS
    æĢ¨
    0.06
     enforce
    0.06
     enforced
    0.06
     enforcement
    0.06
     restrict
    0.06
     žal
    0.06
     boycott
    0.06
     ucfirst
    0.06
    ragen
    0.06
     selectively
    0.05
    Activations Density 0.049%

    No Known Activations