© 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. 12242
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by conjunctions or new clauses

    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
    Negative Logits
    è¾¾åĪ°äºĨ
    -0.10
    æ´ĭæ´ĭ
    -0.09
    çķĻä¸ĭäºĨ
    -0.09
    å®ĮæĪIJäºĨ
    -0.09
    è¿ĺåİŁ
    -0.09
    å±ħ
    -0.09
    åıįèĢĮ
    -0.08
    æľŁæľĽ
    -0.08
    æĮºå¥½
    -0.08
    åıįåĢĴ
    -0.08
    POSITIVE LOGITS
    è¿Ļæł·æīįèĥ½
    0.21
    æīįèĥ½
    0.19
    åIJ¦åĪĻ
    0.19
    æīįè¡Į
    0.19
    åIJ¦åīĩ
    0.16
    æīįèĥ½å¤Ł
    0.15
    æīįåı¯ä»¥
    0.15
     otherwise
    0.14
     lest
    0.13
    ä¸įçĦ¶
    0.13
    Activations Density 0.295%

    No Known Activations