© 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. 17620
    Prev
    Next
    INDEX
    Explanations

    instead of, not x, rather x

    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
    ught
    -0.10
     Tray
    -0.09
    ough
    -0.08
    gif
    -0.08
    æĮĤéĴ©
    -0.08
    èİ·å¾ĹæĽ´å¤ļ
    -0.08
    æĪĸåħ¶ä»ĸ
    -0.08
    bas
    -0.08
    ubby
    -0.07
     Rough
    -0.07
    POSITIVE LOGITS
    äºĮèĢħ
    0.11
    躲éģ¿
    0.11
    好å¥ĩ
    0.10
     Pik
    0.10
    _PAUSE
    0.10
    åīįèĢħ
    0.10
    ä¸İåħ¶
    0.10
    éĢĥ离
    0.09
    ç»ıå¼Ģ
    0.09
    å³»
    0.09
    Activations Density 0.110%

    No Known Activations