© 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. 22161
    Prev
    Next
    INDEX
    Explanations

    code block start/end delimiters

    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
    伶
    -0.07
     vole
    -0.06
    artal
    -0.06
    æijĨäºĨ
    -0.06
    itra
    -0.06
    ä½Ĩåħ¶å®ŀ
    -0.06
    éķľå¤´éĩĮ
    -0.06
    å¹²ç»ĥ
    -0.06
     tunai
    -0.06
    çļĦäºĭæĥħäºĨ
    -0.06
    POSITIVE LOGITS
    ‘
    0.06
    кин
    0.05
     e
    0.05
    iht
    0.05
     getattr
    0.05
     tank
    0.05
    ç©´
    0.05
     pil
    0.05
     prob
    0.05
     stri
    0.05
    Activations Density 0.036%

    No Known Activations