© 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. 52288
    Prev
    Next
    INDEX
    Explanations

    defenses and protections

    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
    åĽŀåij³
    -0.11
    èĽŁ
    -0.10
    åīįéĢĶ
    -0.09
    FSIZE
    -0.08
    ä¸Ŀ绸ä¹ĭè·¯
    -0.08
    éļ¾å¿ĺ
    -0.08
    гин
    -0.08
    èŁł
    -0.08
     Banner
    -0.08
    ouser
    -0.08
    POSITIVE LOGITS
    æݪæĸ½
    0.17
     defenses
    0.14
    éĺ²å¾¡
    0.14
    éĺ²æĬ¤
    0.13
    éĺ²çº¿
    0.13
     barriers
    0.12
     measures
    0.12
     protections
    0.11
    å®īä¿Ŀ
    0.11
    éĺ²èĮĥ
    0.11
    Activations Density 0.109%

    No Known Activations