© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10282
    Prev
    Next
    INDEX
    Explanations

    `into` or `came` followed by movement verbs

    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
    ENTA
    -0.09
    .EventHandler
    -0.09
    åıĭ
    -0.08
    æįĭ
    -0.08
    åľ°åĮº
    -0.08
    æ°ijéĹ´
    -0.08
    座
    -0.08
    =event
    -0.08
    åĮ¹
    -0.08
    ello
    -0.08
    POSITIVE LOGITS
    起身
    0.10
    æĿ¥åĽŀ
    0.09
    zilla
    0.09
    åħ³
    0.09
    ä¼ļåIJĮ
    0.09
    æĮĩçĿĢ
    0.09
    inant
    0.08
     Kitchen
    0.08
    çļĦèĦļæŃ¥
    0.08
    è´¯
    0.08
    Activations Density 0.048%

    No Known Activations