© 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. 30364
    Prev
    Next
    INDEX
    Explanations

    Reverend, Honorable, names

    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
    ä¸Ńå°ı
    -0.10
    åħļåĴĮåĽ½å®¶
    -0.09
    éĩįè¦ģæĦıä¹ī
    -0.09
    æ°Ķæ°Ľ
    -0.09
    sov
    -0.08
    éľĸ
    -0.08
    elp
    -0.08
    åıĤåĬłäºĨ
    -0.08
     Bannon
    -0.08
    ä¸įåİ»
    -0.08
    POSITIVE LOGITS
    esso
    0.09
    æ³ķ人
    0.09
    볨
    0.09
    åݲ
    0.09
    tur
    0.08
    人士
    0.08
    esses
    0.08
    .Blue
    0.08
    esktop
    0.08
    onical
    0.08
    Activations Density 0.011%

    No Known Activations