© 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. 18197
    Prev
    Next
    INDEX
    Explanations

    list item number

    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
    ات
    -0.09
    s
    -0.07
    عادة
    -0.06
    ÙĪÙĦ
    -0.06
    es
    -0.06
    Ñĭ
    -0.06
    ets
    -0.06
    istics
    -0.06
    sache
    -0.06
    ä¸Ń磩
    -0.06
    POSITIVE LOGITS
    ting
    0.07
    orer
    0.07
    tak
    0.07
    antisipasi
    0.07
    Ø©
    0.06
    tte
    0.06
    çIJĨå·¥åѦéĻ¢
    0.06
    aller
    0.06
    ter
    0.06
     Ñģобой
    0.06
    Activations Density 0.043%

    No Known Activations