© 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. 38842
    Prev
    Next
    INDEX
    Explanations

    copy at

    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
    æĹ¢è¦ģ
    -0.08
    è¿Ļç±»
    -0.08
     either
    -0.08
     мало
    -0.08
    å¸ĮæľĽå¤§å®¶
    -0.08
    ÂŃs
    -0.08
    -0.08
    åĴĮä»ĸçļĦ
    -0.08
    æĹłè®ºæĺ¯
    -0.07
    åħĭæĸ¯
    -0.07
    POSITIVE LOGITS
    缴æİ¥
    0.12
    éĢµ
    0.10
    综åIJĪä½ĵ
    0.09
    æīĭä¸ŃçļĦ
    0.09
    æľ¬èº«çļĦ
    0.09
    rand
    0.09
    å¹´ä¹ĭä¹ħ
    0.09
    åħ·ä½ĵåĨħ容
    0.09
    çĻ½æ²Ļ
    0.09
     direct
    0.09
    Activations Density 0.047%

    No Known Activations