© 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. 56090
    Prev
    Next
    INDEX
    Explanations

    review and editorial content

    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
    æĬĬèĩªå·±
    -0.09
    ä»İä¸ļ
    -0.08
    ause
    -0.08
    ds
    -0.08
    åħ¬å¼Ģ
    -0.08
    顾客
    -0.08
    é¢Ĩ
    -0.08
    èĮij
    -0.08
    DER
    -0.08
    éĶĢåĶ®é¢Ŀ
    -0.08
    POSITIVE LOGITS
    æµĭè¯Ħ
    0.11
    è¯Ħæµĭ
    0.11
    åĪĨæľŁ
    0.11
    æĬµ
    0.09
     Editorial
    0.09
    åķľ
    0.09
    æķĻä½ł
    0.08
    estruct
    0.08
    /editor
    0.08
     editorial
    0.08
    Activations Density 0.063%

    No Known Activations