© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37904
    Prev
    Next
    INDEX
    Explanations

    familiar with sources

    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
    åħģ
    -0.09
    ãĥ³ãĤ°
    -0.08
    æį¢æĿ¥
    -0.08
     nice
    -0.08
    éķ¿éĢĶ
    -0.08
     hem
    -0.08
     PROT
    -0.08
    hem
    -0.08
     protester
    -0.08
     Hem
    -0.08
    POSITIVE LOGITS
    æѤäºĭ
    0.10
    OMB
    0.10
    对该
    0.09
    iday
    0.09
    .GetObject
    0.09
    ç½®äºİ
    0.09
    æĹ¥æ¶Īæģ¯
    0.09
    è¯¥é¡¹çĽ®
    0.09
    ç¬ĶèĢħ
    0.09
    ä½ľèĢħæľ¬äºº
    0.09
    Activations Density 0.007%

    No Known Activations