© 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. 19586
    Prev
    Next
    INDEX
    Explanations

    in contrast

    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
    aris
    -0.09
    äºĭåĦ¿
    -0.09
    storybook
    -0.09
    ugu
    -0.09
    horse
    -0.08
    isphere
    -0.08
    æݼ
    -0.08
    楣
    -0.08
    .photos
    -0.08
    æıIJ示
    -0.08
    POSITIVE LOGITS
    ingly
    0.17
    ä¹ĭä¸ĭ
    0.13
    ively
    0.10
    ably
    0.10
    ely
    0.10
    æľīäºĽ
    0.09
    å±ĢéĿ¢
    0.09
    çݰ代社ä¼ļ
    0.09
    äºİ
    0.09
    åħķ
    0.09
    Activations Density 0.017%

    No Known Activations