© 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. 32658
    Prev
    Next
    INDEX
    Explanations

    some of

    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
    éĵ¿
    -0.11
    æIJľéĽĨ
    -0.09
    oll
    -0.09
    鹤
    -0.09
    ä»»ä½ķ
    -0.09
    æĶ¶éĽĨ
    -0.08
    æľīä»»ä½ķ
    -0.08
    æ¹Ľ
    -0.08
    iae
    -0.08
    æĢ§çļĦ
    -0.08
    POSITIVE LOGITS
    perature
    0.10
    ä¹Łè®¸
    0.10
    åĿı
    0.09
    éĿ©
    0.08
     cleric
    0.08
    并ä¸įèĥ½
    0.08
    EventData
    0.08
    æľĢåŁºæľ¬
    0.08
    å£ŀ
    0.08
    没è§ģè¿ĩ
    0.08
    Activations Density 0.057%

    No Known Activations