© 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. 26844
    Prev
    Next
    INDEX
    Explanations

    Purim, Matisse, Hapes, microcks, Timrat

    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
    妹
    -0.10
     Bout
    -0.10
    æĤĶ
    -0.09
    åĩłå®¶
    -0.09
    Pont
    -0.09
    çľĭä¸įåĪ°
    -0.09
    æĮŁ
    -0.09
    ething
    -0.08
    çıŃ车
    -0.08
     GPA
    -0.08
    POSITIVE LOGITS
    æĬ¬å¤´
    0.10
    åĶ®åIJİ
    0.10
    LEAR
    0.10
    aps
    0.09
    etermination
    0.09
    evin
    0.09
     cater
    0.09
    解æĶ¾
    0.09
     ogs
    0.09
    castle
    0.09
    Activations Density 0.075%

    No Known Activations