© 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. 6845
    Prev
    Next
    INDEX
    Explanations

    from these words

    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
    为äºĨæĽ´å¥½
    -0.09
     меÑģÑĤо
    -0.09
    éĹ¼
    -0.09
    á»ĥn
    -0.09
    åĺī
    -0.09
    à¸ģลาà¸ĩ
    -0.08
    ffect
    -0.08
    ä¸įéĢĤ
    -0.08
    æĹłè¾ľ
    -0.08
    .enterprise
    -0.08
    POSITIVE LOGITS
    çĿĢä»ĸ
    0.10
    æŃ¤ä¸¾
    0.09
    å½»
    0.09
    ç»ĪçĶŁ
    0.09
     having
    0.09
    lessly
    0.09
     being
    0.09
    åIJĪä½ľåħ³ç³»
    0.08
    çŁ¥éģĵèĩªå·±
    0.08
    bench
    0.08
    Activations Density 0.028%

    No Known Activations