© 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. 10826
    Prev
    Next
    INDEX
    Explanations

    then followed by nouns

    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
    æľī两ç§į
    -0.10
    ',...↵
    -0.09
    TÄ°
    -0.09
    æľīä¸Ģç§į
    -0.09
    lian
    -0.09
    .sap
    -0.09
    éĹ®é¢ĺæĺ¯
    -0.09
    .openg
    -0.09
    åºĶ注æĦı
    -0.09
    饮
    -0.09
    POSITIVE LOGITS
    ä¹ĥèĩ³
    0.14
    çĶļèĩ³
    0.10
    以åıĬåħ¶ä»ĸ
    0.09
    çĶļèĩ³æĺ¯
    0.09
    缴èĩ³
    0.08
     maybe
    0.08
    继
    0.08
    çŃīçŃī
    0.08
    athers
    0.08
    urer
    0.08
    Activations Density 0.071%

    No Known Activations