© 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. 52589
    Prev
    Next
    INDEX
    Explanations

    general information or context

    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
    irit
    -0.09
    iggins
    -0.09
    åħ¶ä»ĸåĽ½å®¶
    -0.08
    ä¸ĵ项è¡ĮåĬ¨
    -0.08
    [
    -0.08
    æ¿Ģ
    -0.08
    èĪ¸
    -0.08
    è¿Ļ个æĹ¶ä»£
    -0.08
    缸åħ³äººåijĺ
    -0.08
    è§ģæķĪ
    -0.08
    POSITIVE LOGITS
    -purpose
    0.12
    zed
    0.11
    ugen
    0.09
    å®Į
    0.09
    /general
    0.09
    å¨ģ
    0.09
    ities
    0.09
    ä¼Ļ
    0.09
    åĵģçīĮçļĦ
    0.09
    izable
    0.09
    Activations Density 0.035%

    No Known Activations