© 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. 19390
    Prev
    Next
    INDEX
    Explanations

    read or view privileges

    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
     intens
    -0.05
    amac
    -0.05
     spre
    -0.05
    æī¶
    -0.05
    è¿Ľè¡ĮæŶ
    -0.05
    cape
    -0.05
     tens
    -0.05
     ulang
    -0.05
     conven
    -0.05
     thicker
    -0.05
    POSITIVE LOGITS
    åIJĦéłħ
    0.06
    á»ģn
    0.06
    ective
    0.06
    меÑģ
    0.06
    ä¸ĢåĪĩçļĦ
    0.06
    asia
    0.06
    à¹Īà¸Ńà¸ĩ
    0.06
     dreaded
    0.06
    à¸Ĥัà¹īà¸Ļ
    0.06
    اÙģÙĬØ©
    0.06
    Activations Density 0.006%

    No Known Activations