© 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. 19618
    Prev
    Next
    INDEX
    Explanations

    much to

    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
    西éĥ¨
    -0.10
     Rare
    -0.10
    éĢ®
    -0.10
     bears
    -0.09
    åįĥ
    -0.08
    à¸ģา
    -0.08
     SPEC
    -0.08
    æŧ
    -0.08
    SPEC
    -0.08
     ÑģамÑĭе
    -0.08
    POSITIVE LOGITS
    彦
    0.10
    æĦŁåıĹåĪ°äºĨ
    0.10
    åĢ¡
    0.09
    OfType
    0.09
    åļ£
    0.09
    çĻ½è¡£
    0.09
    人éĹ´
    0.09
    线ä¸ĭ
    0.09
    éĴľ
    0.09
    æĸĩåĮĸåĴĮ
    0.09
    Activations Density 0.041%

    No Known Activations