© 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. 57021
    Prev
    Next
    INDEX
    Explanations

    people who create or lead

    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
    ä»Ģ
    -0.09
    æĻļæĬ¥
    -0.09
    lesen
    -0.08
    ohan
    -0.08
    çľĭä¸Ĭåİ»
    -0.08
    ash
    -0.08
     NotImplemented
    -0.08
    è¿ĻéĩĮæĺ¯
    -0.08
    iten
    -0.07
    änd
    -0.07
    POSITIVE LOGITS
     пÑĢоÑģ
    0.10
    çł´äº§
    0.10
    çķĮçļĦ
    0.10
    èĢħçļĦ
    0.10
    大æĪ·
    0.09
    .defaults
    0.09
     Mahm
    0.09
    QL
    0.09
    omer
    0.08
    çĪ±åĽ½ä¸»ä¹ī
    0.08
    Activations Density 0.114%

    No Known Activations