© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 19577
    Prev
    Next
    INDEX
    Explanations

    `a` followed by visualization/report words

    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
     specialised
    -0.06
     specials
    -0.06
     strani
    -0.06
    ç¿»å¼Ģ
    -0.06
    timo
    -0.06
    มี
    -0.06
    çĻĴ
    -0.06
    поло
    -0.06
    ãģ¦
    -0.06
    Regular
    -0.06
    POSITIVE LOGITS
    agni
    0.07
    osite
    0.06
    奴
    0.05
     din
    0.05
    ç¥
    0.05
     dynamic
    0.05
    éĴī
    0.05
    ä¸įæģ¯
    0.05
     dynamically
    0.05
     grand
    0.05
    Activations Density 0.009%

    No Known Activations