© 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. 52729
    Prev
    Next
    INDEX
    Explanations

    filling, full, capacity

    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
    åIJİ代
    -0.10
    ä¸ĭ车
    -0.10
    ç»ıéªĮ丰å¯Į
    -0.09
    ä¿Ŀå§Ĩ
    -0.09
    两åı¥
    -0.09
    AccessType
    -0.09
    .rf
    -0.08
    lcd
    -0.08
    åįĵ
    -0.08
    åĬŀæ³ķ
    -0.08
    POSITIVE LOGITS
    å¡«åħħ
    0.13
    满
    0.13
    åħħåĪĨåĪ©çĶ¨
    0.13
    空
    0.12
     filling
    0.12
    éĹ²ç½®
    0.11
     fill
    0.11
    ä¸į满
    0.11
    filled
    0.11
    滿
    0.11
    Activations Density 0.058%

    No Known Activations