© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3045
    Prev
    Next
    INDEX
    Explanations

    answer and answers

    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
    罪
    -0.09
    éģĩä¸Ĭ
    -0.09
    干货
    -0.09
    èĶ¬
    -0.08
    .Done
    -0.08
    aria
    -0.08
    /fa
    -0.08
    mare
    -0.08
    æłĩå¿Ĺ
    -0.08
    è½®æµģ
    -0.08
    POSITIVE LOGITS
    çŃĶæ¡Ī
    0.16
     answer
    0.15
    çļĦçŃĶæ¡Ī
    0.14
    è¿Ļ个éĹ®é¢ĺ
    0.12
    åĽŀçŃĶ
    0.12
    çļĦåĽŀçŃĶ
    0.12
     answers
    0.12
    ëĭµ
    0.12
    çŃĶ
    0.12
    çŃĶå¤į
    0.11
    Activations Density 0.056%

    No Known Activations