© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 36522
    Prev
    Next
    INDEX
    Explanations

    adhesive, gel, catheter, AcA, Bilicheck, Rely-a-bond, ChiroSolve, lens, Eaziglide, Net

    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
    è¶ħé«ĺ
    -0.09
    çļĩ
    -0.09
    å®Į
    -0.09
    ANTE
    -0.08
    è¿·ä¿¡
    -0.08
    ä½ĵ彩
    -0.08
    ideshow
    -0.08
    使人
    -0.08
    æĴĴ
    -0.08
    OWN
    -0.08
    POSITIVE LOGITS
    ®
    0.11
    âĦ¢
    0.10
    ®,
    0.09
    äºĭåıij
    0.09
    讫
    0.09
    ฯ
    0.09
    âĵĩ
    0.09
    çĭ¸
    0.08
    Allocator
    0.08
    ãĥĵãĥ¥ãĥ¼
    0.08
    Activations Density 0.356%

    No Known Activations