© 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. 50975
    Prev
    Next
    INDEX
    Explanations

    occur in nations

    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
    ä¸ĢåĪĩéĥ½
    -0.12
    ä¸ĢæĶ¯
    -0.11
    è¿ĺæĥ³
    -0.10
    ä¸ĢåĪĩ
    -0.10
     sometimes
    -0.09
    å¤ļæĸ¹
    -0.09
    çŃīå¤ļç§į
    -0.09
    éĥ¨åĪĨåľ°åĮº
    -0.09
    æĢ»æĺ¯
    -0.09
    ä¸Ģ段
    -0.08
    POSITIVE LOGITS
    è¦ģä¹Ī
    0.21
    ä¸įåIJĮç¨ĭ度
    0.17
    either
    0.16
    æĪĸå¤ļæĪĸå°ij
    0.16
     either
    0.16
    Either
    0.14
     Either
    0.14
    æĪĸèĢħæĺ¯
    0.12
    äºĽä»Ģä¹Ī
    0.12
     themselves
    0.11
    Activations Density 0.352%

    No Known Activations