© 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. 28101
    Prev
    Next
    INDEX
    Explanations

    pregnant women

    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
     Maur
    -0.09
    obl
    -0.09
    åijĪ
    -0.09
    è¢ľ
    -0.09
    vers
    -0.08
    ä¸įäºĨ
    -0.08
    dur
    -0.08
    DDD
    -0.08
    aid
    -0.08
    spiel
    -0.08
    POSITIVE LOGITS
    çľĭä»ĸ
    0.09
    èĥĢ
    0.09
    åĩĨ
    0.09
    ening
    0.08
     fancy
    0.08
    èģĨåIJ¬
    0.08
    涯
    0.08
     AO
    0.08
    ç»Ħç»ĩå®ŀæĸ½
    0.08
    åħ¸ç¤¼
    0.08
    Activations Density 0.032%

    No Known Activations