© 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. 21987
    Prev
    Next
    INDEX
    Explanations

    comparative or conditional words

    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
    å®Ļ
    -0.11
    åĤį
    -0.10
    çļĦçģµéŃĤ
    -0.09
    ä¸įåIJ«
    -0.09
    æµĵæµĵçļĦ
    -0.09
    ä¸įåĥı
    -0.09
    çłĶ究ä¸Ńå¿ĥ
    -0.09
    æľĢå¤ļ
    -0.08
    ActionBar
    -0.08
    æľĢåŁºæľ¬çļĦ
    -0.08
    POSITIVE LOGITS
    æľīæĽ´å¥½çļĦ
    0.18
    æĽ´å¥½çļĦ
    0.18
    åħ¶ä»ĸ
    0.17
    æĽ´å¥½
    0.17
    åĪ¥ãģ®
    0.16
    æĽ´éĢĤåIJĪ
    0.16
    åı¦æľī
    0.15
     better
    0.15
    åĪ«çļĦ
    0.15
    åħ¶ä»ĸçļĦ
    0.15
    Activations Density 0.162%

    No Known Activations