© 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. 26566
    Prev
    Next
    INDEX
    Explanations

    acterial ghosts

    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
    à¹Ħร
    -0.09
     ×ij×ĵר×ļ
    -0.08
    core
    -0.08
    注
    -0.08
     Pear
    -0.08
    leftrightarrow
    -0.08
     Peyton
    -0.08
    è¾į
    -0.08
    æ¶ī
    -0.08
    others
    -0.08
    POSITIVE LOGITS
    TF
    0.11
     Excell
    0.09
     TF
    0.09
    åħ·æľīä¸Ģå®ļ
    0.09
    å¤ĸåĽ½
    0.09
    ä¸Ģ页
    0.09
    çĶ·åŃIJ
    0.09
    åħ¨æ°ij
    0.08
    lds
    0.08
    ä¸į容
    0.08
    Activations Density 0.104%

    No Known Activations