© 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. 54613
    Prev
    Next
    INDEX
    Explanations

    in some cases

    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
    é£İæ°Ķ
    -0.10
    æĭĽå¾ħ
    -0.10
    ç§įç§į
    -0.09
    严ç¦ģ
    -0.09
    éĻ¤æѤ
    -0.09
    uum
    -0.09
    æĮºå¥½
    -0.09
    æģģ
    -0.09
    ä¹ĭä½Ļ
    -0.09
    æĦŁåĨĴ
    -0.09
    POSITIVE LOGITS
    çĶļèĩ³æĺ¯
    0.16
    çĶļèĩ³
    0.12
    ités
    0.09
    hed
    0.09
     sogar
    0.09
    æĪIJè´¥
    0.09
    çĶļèĩ³è¿ĺ
    0.09
    orts
    0.09
    è¶ħè¿ĩäºĨ
    0.08
    ä¸Ģ个å¤ļæľĪ
    0.08
    Activations Density 0.050%

    No Known Activations