© 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. 4261
    Prev
    Next
    INDEX
    Explanations

    Community or area followed by a specific institution

    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
    åijĺ
    -0.10
    弯
    -0.09
    éªı
    -0.08
     Stub
    -0.08
    涨
    -0.08
    (TM
    -0.08
    alue
    -0.08
    æ¹¾
    -0.08
    å°¾
    -0.08
    atore
    -0.08
    POSITIVE LOGITS
    ä¸Ńå¤ĸ
    0.10
    åĹ«
    0.09
    åĽ¹
    0.09
    佶
    0.09
    åįļçī©
    0.09
    å¸ĪèĮĥ
    0.09
    åľ¨åĽ½å¤ĸ
    0.09
    好ä¸į好
    0.09
    å°ıå§ijå¨ĺ
    0.09
    æŀģé«ĺ
    0.09
    Activations Density 0.020%

    No Known Activations