© 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. 21088
    Prev
    Next
    INDEX
    Explanations

    two separate

    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
    年建
    -0.06
    มà¸Ļ
    -0.06
    оÑħ
    -0.06
     standardized
    -0.06
    echt
    -0.06
    Ñħло
    -0.05
    çļĦæłĩåĩĨ
    -0.05
    æ°ĶåĴĮ
    -0.05
    ehr
    -0.05
    åĴĮä¿¡æģ¯åĮĸ
    -0.05
    POSITIVE LOGITS
    roy
    0.06
    /dc
    0.06
    å¾®
    0.06
     ÙĨØ´
    0.06
     diges
    0.06
    yen
    0.06
    mue
    0.06
    ypo
    0.06
    缩
    0.05
    جز
    0.05
    Activations Density 0.021%

    No Known Activations