© 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. 27273
    Prev
    Next
    INDEX
    Explanations

    we've, we're, we can, we should

    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
    åĪ¶æľį
    -0.10
    åĬ¨æĦŁ
    -0.09
    ä¸įåIJĮç¨ĭ度
    -0.09
    æĪij们认为
    -0.09
    ä¸įåı¯æĢĿè®®
    -0.08
    çħ¦
    -0.08
    令人
    -0.08
    åıªè¦ģä½ł
    -0.08
    /cop
    -0.08
    宣称
    -0.08
    POSITIVE LOGITS
    ä»ĸèĩªå·±
    0.09
    cover
    0.09
    igh
    0.09
     interven
    0.09
    ainers
    0.09
    .Tool
    0.09
    WG
    0.09
    çłĶ究çĶŁ
    0.08
    åħ¬åħ±èµĦæºIJ
    0.08
    horn
    0.08
    Activations Density 0.056%

    No Known Activations