© 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. 7363
    Prev
    Next
    INDEX
    Explanations

    blog

    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
    .pref
    -0.09
    oba
    -0.09
    åĪĽæĸ°èĥ½åĬĽ
    -0.09
     Remaining
    -0.08
    æ¤IJ
    -0.08
    æ·¡æ·¡çļĦ
    -0.08
    æĢ«
    -0.08
    童年
    -0.08
    åħ¬å¼Ģåıij
    -0.08
     Fle
    -0.08
    POSITIVE LOGITS
    éŨåı£
    0.12
    (blog
    0.10
    /blog
    0.10
    è¡®
    0.10
    槿
    0.10
    ä¹īåĬ¡
    0.09
    åįļ客
    0.09
    inden
    0.09
    åĪ¶
    0.09
    ä¸Ģ线
    0.09
    Activations Density 0.061%

    No Known Activations