© 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. 57002
    Prev
    Next
    INDEX
    Explanations

    repairing damage

    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
    æ¯Ļ
    -0.10
     Pars
    -0.09
    æĻ®æĥł
    -0.09
    íĺľ
    -0.09
     identifiable
    -0.09
    ippers
    -0.09
    é¦ħ
    -0.09
    éĺŁåijĺ们
    -0.08
    å°¼æĸ¯
    -0.08
    赦
    -0.08
    POSITIVE LOGITS
    æįŁä¼¤
    0.11
    çł´æįŁ
    0.10
    åıĹæįŁ
    0.10
    ment
    0.09
     damage
    0.09
    修补
    0.09
    çIJĨå·¥
    0.09
     Damage
    0.09
    æįŁåĿı
    0.09
    utan
    0.09
    Activations Density 0.029%

    No Known Activations