© 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. 30021
    Prev
    Next
    INDEX
    Explanations

    best politicalbest burgerbest riderbest rhythmbest NZbest ofbest protagonistsbest ,best Widebest events

    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
    åĪ¶åĵģ
    -0.08
    æľīçĤ¹
    -0.08
    页
    -0.08
    马åĬĽ
    -0.08
    ÑĢай
    -0.08
    è¦ģ注æĦı
    -0.08
     NAT
    -0.08
     Ñģли
    -0.08
    åIJ¯åıij
    -0.08
    éģĹ
    -0.07
    POSITIVE LOGITS
    åĭ¾
    0.10
    irit
    0.10
    æѤ类
    0.10
    ONUS
    0.09
    lest
    0.09
    itus
    0.09
    ä¸Ģéģĵ
    0.09
    ä¸Ģ件äºĭ
    0.09
    ä½ĵçİ°åľ¨
    0.09
    uru
    0.09
    Activations Density 0.025%

    No Known Activations