© 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. 36720
    Prev
    Next
    INDEX
    Explanations

    internal revenue bureau

    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
    管çIJĨåĪ¶åº¦
    -0.09
    åĪ¶åº¦æĶ¹éĿ©
    -0.09
    人å¿ĥ
    -0.08
    丼
    -0.08
    ataire
    -0.08
    acia
    -0.08
    翳
    -0.08
    çı£
    -0.08
    ç»ıèIJ¥çIJĨ念
    -0.08
    çĶ¬
    -0.08
    POSITIVE LOGITS
     Bureau
    0.13
    æĢ»å±Ģ
    0.12
     bureau
    0.12
    缸åħ³éĥ¨éŨ
    0.10
    gov
    0.10
    å±Ģ
    0.10
     Department
    0.10
    çŃīéĥ¨éŨ
    0.10
     authorities
    0.10
    æľīåħ³éĥ¨éŨ
    0.09
    Activations Density 0.162%

    No Known Activations