© 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. 17165
    Prev
    Next
    INDEX
    Explanations

    need to use, rewrite, or switch

    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
    æµĵæµĵ
    -0.10
    uran
    -0.10
    .endpoint
    -0.09
    å·ħ
    -0.08
    æķ´åIJĪ
    -0.08
    éĺª
    -0.08
    ä»ĭ
    -0.08
    oir
    -0.08
    æĹ¢æľī
    -0.08
     Chapel
    -0.08
    POSITIVE LOGITS
    è¦ģä¹Ī
    0.16
    å¹²èĦĨ
    0.12
    åĪ¥ãģ®
    0.11
    åĩºåĽ½
    0.11
    åı¦è¡Į
    0.11
    æī¾ä¸ª
    0.10
    æļĤæŶ
    0.10
    ç¥Ī祷
    0.10
    åĪĽåĬŀ
    0.10
    åħ¶ä»ĸçļĦ
    0.10
    Activations Density 0.198%

    No Known Activations