© 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. 33720
    Prev
    Next
    INDEX
    Explanations

    too much or more

    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
    åĵªéĩĮ
    -0.09
    大æ¦Ĥæĺ¯
    -0.09
    ç¨įç¨į
    -0.09
    大æ¦Ĥ
    -0.09
    -era
    -0.08
    æĢİä¹Īæł·
    -0.08
    æŀľ
    -0.08
     Willis
    -0.08
    èĬ±æľµ
    -0.08
    å·®ä¸įå¤ļ
    -0.08
    POSITIVE LOGITS
    ä¸įçĦ¶
    0.10
    äºijéĽĨ
    0.09
    _DS
    0.09
    åıĺæĢģ
    0.09
    mland
    0.09
    isque
    0.09
    llib
    0.09
    缸è§ģ
    0.09
    èĢĮæĪIJ
    0.09
    à¸ŀฤ
    0.08
    Activations Density 0.012%

    No Known Activations