© 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. 26058
    Prev
    Next
    INDEX
    Explanations

    pot will be filled

    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
     recip
    -0.09
     REC
    -0.09
     mini
    -0.09
    åıĽ
    -0.08
    亲
    -0.08
    亲è¿ij
    -0.08
    äºĴ缸
    -0.08
    çĶµåĻ¨
    -0.08
    ä½ĵè´´
    -0.08
    éļ¼
    -0.08
    POSITIVE LOGITS
    çŀ½
    0.10
    zie
    0.10
    éļĶ
    0.09
    æ°´ä¸Ń
    0.09
    çħ®
    0.09
    à¸Ńะ
    0.09
    åľ¨å¸Ĥåľºä¸Ĭ
    0.09
    åħĭæĸ¯
    0.09
    æ°´éĩı
    0.09
    uga
    0.09
    Activations Density 0.135%

    No Known Activations