© 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. 56157
    Prev
    Next
    INDEX
    Explanations

    extraction operators

    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
    为äºĨ
    -0.10
    è²Į
    -0.09
    ayer
    -0.09
     cw
    -0.09
     èĭ¥è¦ģ
    -0.08
    为äºĨ让
    -0.08
    ertificate
    -0.08
    æı¡
    -0.08
    åĩºäºİ
    -0.08
    íķĦ
    -0.08
    POSITIVE LOGITS
    åĬłä»¥
    0.10
    åģı
    0.10
    çĺł
    0.10
    赶åĪ°
    0.10
    äºĪ以
    0.10
    åĴĮæĬĢæľ¯
    0.10
    èݼ
    0.09
    èµĦ
    0.09
    æıIJä¾ĽçļĦ
    0.09
    åģıåIJij
    0.09
    Activations Density 0.661%

    No Known Activations