© 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. 35983
    Prev
    Next
    INDEX
    Explanations

    true, indeed

    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
    æijĩ
    -0.10
    udo
    -0.10
     allegedly
    -0.09
    æīĢè°ĵçļĦ
    -0.09
    æ²±
    -0.09
    éļ»
    -0.09
    强åĪ¶
    -0.09
    ä¿¡èµĸ
    -0.09
    绶
    -0.09
     оÑĦоÑĢм
    -0.09
    POSITIVE LOGITS
    ç¡®å®ŀæĺ¯
    0.11
    çļĦç¡®æĺ¯
    0.10
     indeed
    0.10
    没éĶĻ
    0.10
    ç²¾çĽĬæ±Ĥç²¾
    0.09
    ippo
    0.09
    기ëıĦ
    0.09
    ç¡®å®ŀ
    0.09
    ç²¾çĽĬ
    0.09
    屿
    0.09
    Activations Density 0.058%

    No Known Activations