© 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. 50006
    Prev
    Next
    INDEX
    Explanations

    lack or absence of

    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
     Buckley
    -0.09
    两次
    -0.09
    atte
    -0.09
    IDA
    -0.08
    ä»İäºĭ
    -0.08
     пÑĢож
    -0.08
    çĶŁæ´»ä¹łæĥ¯
    -0.08
    rams
    -0.08
    áĥ§
    -0.08
    å®ŀåľ¨
    -0.08
    POSITIVE LOGITS
    éĹ®è´£
    0.11
    æĬµæĬĹåĬĽ
    0.11
    æĦıè¯ĨåĪ°
    0.10
    ä¸ĢçĤ¹åĦ¿
    0.10
    çĽij管
    0.10
    aven
    0.09
    ä¸Ģå¥Ĺ
    0.09
    ä¸Ģæľ¬ä¹¦
    0.09
    åĬłä¹ĭ
    0.09
    åĩıå¼±
    0.09
    Activations Density 0.139%

    No Known Activations