© 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. 36917
    Prev
    Next
    INDEX
    Explanations

    will be / will have

    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
    åı¯ä»¥æĽ´å¥½
    -0.11
     полÑĮзÑĥ
    -0.11
    elite
    -0.11
    è°ģçŁ¥éģĵ
    -0.10
    steen
    -0.10
    ç¶ļ
    -0.10
    èĥ½åľ¨
    -0.10
    gorit
    -0.10
    é©
    -0.10
    éªĦ
    -0.09
    POSITIVE LOGITS
    åĩĨ
    0.09
    å¥ī
    0.09
    "
    0.08
    ä¸ĵå±ŀ
    0.08
     option
    0.08
     IQ
    0.08
    ä¼ı
    0.08
    ä¸ĵçĶ¨
    0.07
    éĵº
    0.07
    ä¾ĭè¡Į
    0.07
    Activations Density 0.066%

    No Known Activations