© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21088
    Prev
    Next
    INDEX
    Explanations

    unfortunately

    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
    æµ®åĬ¨
    -0.08
    æĹłéľĢ
    -0.08
    åħ³éĶ®æĺ¯
    -0.08
    oes
    -0.08
    Challenge
    -0.08
    å¤ĩ
    -0.08
     ÑģÑĥ
    -0.08
    hs
    -0.08
    好åIJĹ
    -0.08
    اÙĨÙĬ
    -0.08
    POSITIVE LOGITS
    ä¹Łåıªèĥ½
    0.12
    Unfortunately
    0.11
    ä¹Łä¸įèĥ½
    0.11
     unfortunately
    0.11
     Unfortunately
    0.10
    åıªèĥ½
    0.10
     sadly
    0.10
    è¿Ļ群
    0.10
     trag
    0.09
    ä¸į幸
    0.09
    Activations Density 0.075%

    No Known Activations