© 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. 18754
    Prev
    Next
    INDEX
    Explanations

    little to no effect

    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
    ç¿ħ
    -0.10
    .grad
    -0.09
    æīįåı¯ä»¥
    -0.09
    iane
    -0.09
    Ñĥни
    -0.09
    ä¸ĢéĿ¢
    -0.08
     ÑģÑĦ
    -0.08
    åŁºè°ĥ
    -0.08
    ä¸įéĢĤåIJĪ
    -0.08
    èįīæ¡Ī
    -0.08
    POSITIVE LOGITS
    羣æŃ£
    0.13
    羣æŃ£çļĦ
    0.13
    æľīçĶ¨
    0.11
    解åĨ³éĹ®é¢ĺ
    0.11
    æľīæĦıä¹ī
    0.10
    æķĪæŀľ
    0.10
    解åĨ³
    0.10
    ä½ľçĶ¨
    0.09
     benefit
    0.09
    å¢ŀè¿Ľ
    0.09
    Activations Density 0.115%

    No Known Activations