© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 26333
    Prev
    Next
    INDEX
    Explanations

    unique identifiers

    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
    ysi
    -0.09
    ÑĦÑĢ
    -0.09
    .ali
    -0.09
    .Direct
    -0.08
    TEL
    -0.08
    legg
    -0.08
    éĶ´
    -0.08
    æĬ½æ£Ģ
    -0.08
    æŶéĹ´ä¸º
    -0.08
    _budget
    -0.08
    POSITIVE LOGITS
    çĭ¬ä¸ĢæĹłäºĮ
    0.15
     unique
    0.14
    çĭ¬çī¹
    0.13
    åĶ¯ä¸ĢçļĦ
    0.13
    _unique
    0.13
     uniq
    0.13
     Ñĥника
    0.12
     Unique
    0.12
    çĭ¬çī¹çļĦ
    0.12
    çļĦçĭ¬çī¹
    0.12
    Activations Density 0.203%

    No Known Activations