© 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. 18112
    Prev
    Next
    INDEX
    Explanations

    structural data fields or columns

    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
    nder
    -0.06
    ÏīÏĤ
    -0.05
    alle
    -0.05
    CA
    -0.05
    alles
    -0.05
    sir
    -0.05
    rosis
    -0.05
    unders
    -0.05
    é¦Ħ饨
    -0.05
    è¿ĺæĺ¯æľī
    -0.05
    POSITIVE LOGITS
     akin
    0.07
    alice
    0.06
    iete
    0.06
    è¨Ģä¹ĭ
    0.06
    ayah
    0.06
    inery
    0.06
     reminiscent
    0.06
    æ¥ŃåĭĻ
    0.06
    æĻ¦
    0.06
    pired
    0.06
    Activations Density 1.700%

    No Known Activations