© 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. 32568
    Prev
    Next
    INDEX
    Explanations

    why does

    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
    advance
    -0.08
    ess
    -0.08
    esses
    -0.08
    ä¸Ģ个人
    -0.07
     super
    -0.07
    å£°éŁ³
    -0.07
    èĻ½çĦ¶
    -0.07
    åĢĴ
    -0.07
     whatever
    -0.07
    çļĦåIJĪä½ľ
    -0.07
    POSITIVE LOGITS
    è¿Ļç§įæĥħåĨµ
    0.17
    è¿Ļéĥ¨åĪĨ
    0.12
    æĺ¯æĪijçļĦ
    0.11
    对æѤ
    0.10
    请éĹ®
    0.10
    æ±²
    0.10
    çļĨãģķãĤĵ
    0.10
    _pdata
    0.10
    é¹Ĺ
    0.10
    æĢİä¹ĪåĽŀäºĭ
    0.10
    Activations Density 0.340%

    No Known Activations