© 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. 34082
    Prev
    Next
    INDEX
    Explanations

    quality and severity

    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¸į失
    -0.10
    è¯ĺ
    -0.09
    绽
    -0.09
    æĬĦ
    -0.09
    ERIC
    -0.09
    çĶļèĩ³
    -0.09
    导
    -0.08
    æľĢ大
    -0.08
    è¾ĥ大çļĦ
    -0.08
    çŃĸ
    -0.08
    POSITIVE LOGITS
    çļĦè´¨éĩı
    0.16
    è´¨éĩı
    0.14
    質éĩı
    0.14
     quality
    0.13
    quality
    0.13
     каÑĩеÑģÑĤво
    0.13
    è´¨éĩıéĹ®é¢ĺ
    0.11
     Quality
    0.11
    Quality
    0.11
    è´¨éĩıåĴĮ
    0.11
    Activations Density 0.372%

    No Known Activations