© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21269
    Prev
    Next
    INDEX
    Explanations

    explaining things is necessary

    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
    åĵī
    -0.09
    nbsp
    -0.08
    è¿Ļä¹Ł
    -0.08
    è¿IJ
    -0.08
    æĹ¢çĦ¶
    -0.08
     Bradley
    -0.08
     ìļ´
    -0.08
    usz
    -0.08
    CUS
    -0.08
     discre
    -0.08
    POSITIVE LOGITS
    å½ĵå±Ģ
    0.09
    opia
    0.09
    åĢ¾
    0.09
     Sung
    0.08
    è¯ħ
    0.08
    culate
    0.08
    ç¾İçļĦ
    0.08
    æĢ§åĴĮ
    0.08
    æĮĸ
    0.08
    aris
    0.08
    Activations Density 0.055%

    No Known Activations