© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 47624
    Prev
    Next
    INDEX
    Explanations

    very/quite/extremely followed by positive adjectives

    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
    Negative Logits
    dy
    -0.10
    衬
    -0.08
    werp
    -0.08
    ourd
    -0.08
    ecture
    -0.08
     kicks
    -0.07
    æŁĵ
    -0.07
    IID
    -0.07
    år
    -0.07
    igator
    -0.07
    POSITIVE LOGITS
     revolver
    0.09
    åѤåĦ¿
    0.09
    ноÑģÑĤ
    0.09
     adam
    0.09
    尤为éĩįè¦ģ
    0.09
    ä¼ģä¸ļæıIJä¾Ľ
    0.09
    伸åĩº
    0.08
    她æĺ¯
    0.08
    UnityEngine
    0.08
    ç°¡åįĺãģ«
    0.08
    Activations Density 0.077%

    No Known Activations