© 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. 2509
    Prev
    Next
    INDEX
    Explanations

    what followed by question words

    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
    æĻ¯å¾·
    -0.10
    Extras
    -0.10
    ä¼ijåģĩ
    -0.10
    åĬŀäºĭ
    -0.10
     Overflow
    -0.09
     اÙĦÙĤض
    -0.09
    rut
    -0.09
    大象
    -0.09
    overflow
    -0.09
    è¿Ļä¸ĢçĤ¹
    -0.09
    POSITIVE LOGITS
    æł·çļĦ
    0.11
    好å¤Ħ
    0.10
    forth
    0.10
    pon
    0.09
    ippers
    0.09
     role
    0.08
    顺çĿĢ
    0.08
    è§Ĵèī²
    0.08
     additional
    0.08
    éļľç¢į
    0.08
    Activations Density 0.056%

    No Known Activations