© 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. 34549
    Prev
    Next
    INDEX
    Explanations

    ownership and customization

    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
    åı¯æĢľ
    -0.09
    çĻ»
    -0.08
     Tricks
    -0.08
    èµ°ä¸Ĭ
    -0.08
    ç¼ij
    -0.07
    çĹĽèĭ¦
    -0.07
    çĪ¬
    -0.07
    楼
    -0.07
    wis
    -0.07
    æĬĺ
    -0.07
    POSITIVE LOGITS
    åĩºä¼Ĺ
    0.12
    ä¸»åľº
    0.10
    客æĪ·çļĦ
    0.10
    åIJĮå¿ĥ
    0.10
    ewe
    0.10
    è°Ľ
    0.10
    交æ±ĩ
    0.10
    æĹħéĢĶ
    0.09
    åĽłåľ°
    0.09
    åĵģçīĮçļĦ
    0.09
    Activations Density 0.294%

    No Known Activations