© 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. 21550
    Prev
    Next
    INDEX
    Explanations

    dashing, daisy, duffle, dainty, daniel, dined, dews, dairying, drenched

    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
    æĭĶ
    -0.10
    å¤ļæķ°
    -0.10
    åīĬåĩı
    -0.09
    æĸ«
    -0.09
    Anime
    -0.09
    ä¸Ńåħ³
    -0.09
     Schmidt
    -0.09
    owi
    -0.09
    ç»Ī身
    -0.09
    褰
    -0.09
    POSITIVE LOGITS
    åľ°ä¸Ĭ
    0.11
    /d
    0.09
    èĮħ
    0.08
    ëĭ´
    0.08
    èĬĴ
    0.08
    hem
    0.08
    说åĪ°è¿ĻéĩĮ
    0.08
    InSection
    0.08
    ture
    0.07
     Kay
    0.07
    Activations Density 0.032%

    No Known Activations