© 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. 26598
    Prev
    Next
    INDEX
    Explanations

    project-related contexts

    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
    åĹŁ
    -0.11
    æĴ¼
    -0.10
    åħ¬å¼Ģåıij
    -0.10
    iro
    -0.10
    çľĭçĤ¹
    -0.10
    æ¿ĢçĥĪ
    -0.09
    è´·
    -0.09
    idis
    -0.09
    å·²æľī
    -0.09
    éĽ¯
    -0.09
    POSITIVE LOGITS
    窮
    0.10
    æĻ®
    0.09
    ç©·
    0.09
    俨
    0.09
    å½ķ
    0.09
     Newtown
    0.08
    éħŀ
    0.08
    æģ»
    0.08
    马ä¸ģ
    0.08
    æĪijçĪ±ä½ł
    0.08
    Activations Density 0.032%

    No Known Activations