© 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. 31124
    Prev
    Next
    INDEX
    Explanations

    official sources, policy, reports, confirmation

    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
    åīįæīĢ
    -0.12
    -être
    -0.09
    /if
    -0.09
    ominated
    -0.09
    /be
    -0.09
    è·³åĩº
    -0.09
    ä¼ģä¸ļåĴĮ
    -0.08
    -hop
    -0.08
    ettle
    -0.08
    è´¨éĩıåĴĮ
    -0.08
    POSITIVE LOGITS
    ity
    0.14
    æ¸łéģĵ
    0.10
    ÙĩÙĦ
    0.10
    ism
    0.09
    æĿ¥è¯´
    0.09
    /math
    0.09
    dehyde
    0.09
    dom
    0.09
    agan
    0.09
    izing
    0.09
    Activations Density 0.013%

    No Known Activations