© 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. 35259
    Prev
    Next
    INDEX
    Explanations

    "would" followed by verbs

    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
    ä¼¼ä¹İæĺ¯
    -0.10
    nee
    -0.09
    人éĥ½
    -0.09
    ä¸įä½ı
    -0.08
    Cho
    -0.08
    èī²ç´ł
    -0.08
     Cho
    -0.08
    ä¸įä½Ĩ
    -0.08
    æĪijå¾Ī
    -0.08
    º
    -0.08
    POSITIVE LOGITS
    è´½
    0.10
    èģĶç³»
    0.09
    Ñīин
    0.09
    illon
    0.08
    è¶ģ
    0.08
    ÇĤ
    0.08
    olg
    0.08
    åĴ¨è¯¢
    0.08
    вин
    0.08
    èļĢ
    0.08
    Activations Density 0.027%

    No Known Activations