© 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. 20217
    Prev
    Next
    INDEX
    Explanations

    reasoning clauses

    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
    ÛĮار
    -0.06
    eer
    -0.06
    iak
    -0.06
    åĵģçļĦ
    -0.06
    ful
    -0.06
    çİĭåŃIJ
    -0.06
    mittel
    -0.06
    и
    -0.06
    hood
    -0.06
    Ùħائة
    -0.06
    POSITIVE LOGITS
    ÑģÑı
    0.06
    combe
    0.06
    abwe
    0.06
    onese
    0.06
    ìĦľ
    0.06
    ictured
    0.05
    ↵     ↵
    0.05
    лдÑĭ
    0.05
    erschein
    0.05
    IgnoreCase
    0.05
    Activations Density 0.002%

    No Known Activations