© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7479
    Prev
    Next
    INDEX
    Explanations

    with relative clauses

    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
    æīĢåľ¨çļĦ
    -0.12
    éĤ£ç§į
    -0.11
    è¿ĻéĩĮçļĦ
    -0.11
    å½ĵåľ°çļĦ
    -0.10
    éĤ£ä¸ªæŶåĢĻ
    -0.10
    ä¹ĭåīįçļĦ
    -0.10
    çľ¼åīįçļĦ
    -0.10
    åģļ人
    -0.09
    æľīä½ķ
    -0.09
    侪
    -0.09
    POSITIVE LOGITS
     whose
    0.09
    whose
    0.09
     which
    0.08
    meer
    0.08
     no
    0.08
    åıĮè¾¹
    0.08
    gel
    0.08
     pit
    0.08
    éĩįè¦ģæĦıä¹ī
    0.08
    è¾ĥé«ĺ
    0.08
    Activations Density 0.218%

    No Known Activations