© 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. 10243
    Prev
    Next
    INDEX
    Explanations

    bookmarked pages or book width

    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
     sack
    -0.11
    burgh
    -0.10
    æĺ¯ä¸Ģ件
    -0.09
    çıī
    -0.09
    鸵
    -0.09
    åıĺæĢģ
    -0.09
    麻çĥ¦
    -0.09
     parity
    -0.09
    åĪ·æĸ°
    -0.09
    ilian
    -0.08
    POSITIVE LOGITS
    页
    0.09
    èµł
    0.09
     margin
    0.09
    æıĴ
    0.09
    éļıå¤Ħ
    0.09
    .setDefault
    0.09
    é¢Ĭ
    0.09
    éĥijéĩį
    0.09
    Bookmark
    0.09
    è¿Ļæľ¬ä¹¦
    0.08
    Activations Density 0.072%

    No Known Activations