© 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. 34932
    Prev
    Next
    INDEX
    Explanations

    rehabilitation and upgrades

    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
    峡
    -0.09
    mith
    -0.08
    åīįæīĢæľªæľī
    -0.08
    æݨéĶĢ
    -0.08
    zas
    -0.08
    aired
    -0.08
    nad
    -0.08
    èŀ
    -0.08
    é¦ĸåıij
    -0.08
    nio
    -0.08
    POSITIVE LOGITS
    æĶ¹éĢł
    0.26
    åįĩ级æĶ¹éĢł
    0.21
    æĶ¹è£ħ
    0.18
    åįĩ级
    0.17
    æĶ¹åĪ¶
    0.16
    æķ´æĶ¹
    0.16
     upgrades
    0.15
    æĶ¹å»º
    0.15
     upgrade
    0.14
    æĶ¹èī¯
    0.14
    Activations Density 0.110%

    No Known Activations