© 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. 26798
    Prev
    Next
    INDEX
    Explanations

    and followed by a noun

    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
    å®ŀå®ŀåľ¨åľ¨
    -0.09
    女æĸ¹
    -0.09
    çĶļèĩ³æĺ¯
    -0.08
    å¹²èĦĨ
    -0.08
    obic
    -0.08
    åĴĮ个人
    -0.08
    çī¹å®ļ
    -0.08
    åIJĪè§Ħ
    -0.08
    åĪ©äºİ
    -0.08
    ä¹ĭ称
    -0.08
    POSITIVE LOGITS
     Gone
    0.10
    å¾ĹåĪ°äºĨ
    0.09
    åĽŀæĥ³
    0.08
    ronic
    0.08
    enville
    0.08
    itra
    0.08
    飧
    0.08
    gone
    0.08
    çļĦ人éĥ½
    0.08
    佩
    0.08
    Activations Density 0.201%

    No Known Activations