© 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. 56230
    Prev
    Next
    INDEX
    Explanations

    incubated / removed / saturated / frozen

    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
    èĥĮä¸Ĭ
    -0.09
    ä¸ĢçĤ¹åĦ¿
    -0.09
    éĢłç¦ı
    -0.08
    磸
    -0.08
     ste
    -0.08
    èĥ¯
    -0.08
    hei
    -0.08
    ç¼ĩ
    -0.07
     embar
    -0.07
    路人
    -0.07
    POSITIVE LOGITS
    ç½®äºİ
    0.10
    ä¸Ĭä¸ĭ游
    0.10
    åIJ«
    0.09
    holders
    0.09
    ombat
    0.09
    æĶ¾ç½®
    0.09
    amt
    0.08
    Ñģал
    0.08
     overnight
    0.08
    è¾±
    0.08
    Activations Density 0.654%

    No Known Activations