© 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. 24126
    Prev
    Next
    INDEX
    Explanations

    awesome

    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
    Negative Logits
    ways
    -0.10
    ãģĦãģ¾ãģĻ
    -0.10
    TING
    -0.10
    noch
    -0.09
    ivity
    -0.09
    zelf
    -0.09
    ss
    -0.09
    west
    -0.09
    б
    -0.09
    rate
    -0.09
    POSITIVE LOGITS
    arness
    0.11
    egis
    0.10
    viz
    0.10
    alse
    0.10
    ÑģÑĤеÑĢ
    0.09
    ething
    0.09
    åħ¶ä¸ŃåĮħæĭ¬
    0.09
    ment
    0.09
    lop
    0.09
    ously
    0.08
    Activations Density 0.005%

    No Known Activations