© 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. 61588
    Prev
    Next
    INDEX
    Explanations

    geographic areas

    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
    (Locale
    -0.09
    åĤħ
    -0.09
    åĬŁå¾·
    -0.09
    afort
    -0.09
     ÑĥÑģи
    -0.09
    å±Ĥåĩºä¸įç©·
    -0.09
    åĢĻéĢī
    -0.09
    çĤ¹
    -0.09
    çĤ¹äºĨ
    -0.09
    çŀ½
    -0.09
    POSITIVE LOGITS
    ä¸Ģéĥ¨åĪĨ
    0.11
    çļĦéĥ¨åĪĨ
    0.11
    çļĦä¸Ģéĥ¨åĪĨ
    0.10
    uco
    0.10
    åĽ´ç»ķ
    0.09
    大åıij
    0.09
     extension
    0.09
     extensions
    0.09
    诸
    0.08
    延伸
    0.08
    Activations Density 0.134%

    No Known Activations