© 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. 30627
    Prev
    Next
    INDEX
    Explanations

    one size fits all

    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
    ric
    -0.10
    åĵī
    -0.10
    æĻĭ
    -0.10
    èıĢ
    -0.09
    adi
    -0.09
    广éĺĶ
    -0.08
    клад
    -0.08
     оÑĩ
    -0.08
    æµ·æ£ł
    -0.08
    .nr
    -0.08
    POSITIVE LOGITS
    åĵį
    0.09
    dden
    0.09
    åĽºå®ļçļĦ
    0.09
    ãģĿãģ®ãģ¾ãģ¾
    0.08
     static
    0.08
     %[
    0.08
    çīĩéĿ¢
    0.08
     blinded
    0.08
    ä»İåīį
    0.08
    æłª
    0.08
    Activations Density 0.043%

    No Known Activations