© 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. 33926
    Prev
    Next
    INDEX
    Explanations

    locations where

    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
    æĿ¥è¿ĻéĩĮ
    -0.12
    åľ¨è¿Ļç§į
    -0.10
    è¿ĻéĩĮæĺ¯
    -0.09
    æĺ¯ä¸ŃåĽ½
    -0.09
    ission
    -0.09
    è¿Ļæĺ¯æĪij们
    -0.09
    è¦ģåľ¨
    -0.09
    åľ¨å¤ĸéĿ¢
    -0.09
    å¤ĸåĩº
    -0.09
    è¿ĻéĩĮéĿ¢
    -0.09
    POSITIVE LOGITS
    åģľçķĻ
    0.11
    ä¸ĢæĬĬ
    0.09
     Saul
    0.08
    çĥ¹
    0.08
    åı¯èĥ½åıijçĶŁ
    0.08
    upon
    0.08
     economy
    0.08
    ħĮ
    0.08
    urai
    0.08
    äºĨå¤ļå°ij
    0.08
    Activations Density 0.066%

    No Known Activations