© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21308
    Prev
    Next
    INDEX
    Explanations

    include, including, includes

    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
    ãģĭãĤĮ
    -0.09
    æĻ·
    -0.09
    åĪ¶æŃ¢
    -0.09
    è¿ĩåĪĨ
    -0.08
    strict
    -0.08
    禺
    -0.08
    大大æıIJé«ĺ
    -0.08
    çĶŁäºİ
    -0.08
     fins
    -0.08
    licable
    -0.08
    POSITIVE LOGITS
    åĮħæĭ¬
    0.23
    åħ¶ä¸ŃåĮħæĭ¬
    0.19
     including
    0.16
    è¿ĺåĮħæĭ¬
    0.16
     includes
    0.16
     include
    0.16
     includ
    0.16
    主è¦ģåĮħæĭ¬
    0.15
    including
    0.14
     Includes
    0.14
    Activations Density 0.135%

    No Known Activations