© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8442
    Prev
    Next
    INDEX
    Explanations

    developed and used

    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
    åij¨æľ«
    -0.10
    对该
    -0.10
    irl
    -0.09
    ccount
    -0.09
    以åıĬåħ¶ä»ĸ
    -0.09
    èĻ¢
    -0.09
    çļĦæĪIJåĬŁ
    -0.09
    åĬ¨éĿĻ
    -0.09
    éĽĨåĽ¢èĤ¡ä»½
    -0.09
    çŃīå¥ĸ
    -0.09
    POSITIVE LOGITS
    ed
    0.10
    æĪĸèĢħ说
    0.09
    inis
    0.09
    /or
    0.08
    lo
    0.08
    irus
    0.08
    edBy
    0.08
    imus
    0.08
    ç¥ŀ
    0.08
    elong
    0.08
    Activations Density 0.164%

    No Known Activations