© 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. 32292
    Prev
    Next
    INDEX
    Explanations

    correcting errors

    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
    åį«è§Ĩ
    -0.10
     Coleman
    -0.09
    holm
    -0.09
    æĪĸå¤ļ
    -0.09
    äºĮæīĭ
    -0.09
    hiba
    -0.09
    åĽŀæĿ¥äºĨ
    -0.08
    åĽŀé¦Ī
    -0.08
    åĽŀæĿ¥
    -0.08
    åıĺå¼Ĥ
    -0.08
    POSITIVE LOGITS
    banks
    0.09
     BAT
    0.09
    饵
    0.09
    两侧
    0.08
    BAT
    0.08
     informal
    0.08
    OrCreate
    0.08
    åħ¬çĽĬæĢ§
    0.08
    ORK
    0.08
    hte
    0.08
    Activations Density 0.007%

    No Known Activations