© 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. 2525
    Prev
    Next
    INDEX
    Explanations

    brushing away or off

    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
    沸
    -0.09
    èģĮä¸ļæĬĢèĥ½
    -0.09
    åı·åı¬
    -0.09
    faces
    -0.09
     Bent
    -0.08
     SOCK
    -0.08
    辫
    -0.08
    è°ĭ
    -0.08
    inker
    -0.08
    chain
    -0.08
    POSITIVE LOGITS
    çĶ©
    0.11
    æݨå¼Ģ
    0.10
    æijĩ
    0.10
    æİī
    0.09
    jal
    0.09
    åĽ½å®¶éĺŁ
    0.09
    -fly
    0.09
    çĿ¬
    0.08
    ä¸į解
    0.08
    .stack
    0.08
    Activations Density 0.023%

    No Known Activations