© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 32928
    Prev
    Next
    INDEX
    Explanations

    taken to

    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
    èĤ²åĦ¿
    -0.10
    æ³¼
    -0.09
    人人
    -0.09
    miss
    -0.09
    æ¡IJ
    -0.09
    äu
    -0.09
    éĻĮ
    -0.09
    çĥ¹
    -0.08
    æŃ»è§Ĵ
    -0.08
    æ²±
    -0.08
    POSITIVE LOGITS
    åīįå¾Ģ
    0.14
    åİ»åĮ»éĻ¢
    0.12
    åıĬæŶ
    0.11
    éĢģåĪ°
    0.11
    éĢģåİ»
    0.10
    azy
    0.10
    åĵªè£¡
    0.09
    æĿ¥åĪ°
    0.09
    åĽŀ家
    0.09
    å°±è¿ij
    0.09
    Activations Density 0.048%

    No Known Activations