© 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. 48789
    Prev
    Next
    INDEX
    Explanations

    on top of

    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
    大åŀĭ
    -0.09
     пÑĢоб
    -0.09
    etal
    -0.08
    åī¯ä½ľçĶ¨
    -0.08
    ä¾Ŀçħ§
    -0.08
    ulling
    -0.08
    伫
    -0.08
    为åĩĨ
    -0.08
    æݢ索
    -0.08
    Fcn
    -0.08
    POSITIVE LOGITS
    åĨįåĬłä¸Ĭ
    0.12
    éĤĦè¦ģ
    0.11
    è¿ĺæĥ³
    0.10
    ä¸įä»ħå¦ĤæѤ
    0.10
    è¿ĺè¦ģ
    0.10
    è¿ĺæľīä¸Ģ个
    0.10
    ä¹ĭä¸Ĭ
    0.10
    è¿ĺæľī
    0.10
    è¿ĺå°Ĩ
    0.10
    è¿ĺå¾Ĺ
    0.09
    Activations Density 0.004%

    No Known Activations