© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 49962
    Prev
    Next
    INDEX
    Explanations

    unique features

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    åħļçļĦ建设
    -0.10
    uled
    -0.10
    arin
    -0.08
    ä¸ĸ人
    -0.08
    eson
    -0.08
    ews
    -0.08
    .gov
    -0.08
    owing
    -0.08
    è°Ĩ
    -0.08
    ouse
    -0.08
    POSITIVE LOGITS
    ä¹ĭå¤Ħ
    0.12
    è§ģ解
    0.11
    åĨįéĢł
    0.09
    设计çIJĨ念
    0.09
    OfWork
    0.09
     finder
    0.09
    è§ĩ
    0.09
    宪
    0.09
    idade
    0.08
    æĸ°é¢ĸ
    0.08
    Activations Density 0.019%

    No Known Activations