© 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. 58879
    Prev
    Next
    INDEX
    Explanations

    grave tone, mailboxes, beep, Foundation claims

    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
     Micha
    -0.09
    elm
    -0.08
    è¸ħ
    -0.08
    以åıĬåħ¶ä»ĸ
    -0.08
     Hun
    -0.07
    æĤĴ
    -0.07
    离å¼ĢäºĨ
    -0.07
    æ·¦
    -0.07
     Kem
    -0.07
     squirrel
    -0.07
    POSITIVE LOGITS
    ä¼¼çļĦ
    0.11
    仿
    0.10
    éķ°
    0.10
    çŃ¾
    0.10
    æĥħèĬĤ
    0.10
    ndl
    0.09
    åĴ¿
    0.09
    åĥŃ
    0.09
    æĥħå¢ĥ
    0.09
     Fort
    0.09
    Activations Density 0.118%

    No Known Activations