© 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. 51408
    Prev
    Next
    INDEX
    Explanations

    unexpected dangers or problems

    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
    çĭ©
    -0.08
    å¿ĥæĦ¿
    -0.08
    æĮĩ导ä¸ĭ
    -0.08
     whipping
    -0.08
    åı¯ä¾Ľ
    -0.08
    å«ģ
    -0.07
    ÙĦاÙħ
    -0.07
    OID
    -0.07
    è°ĥ
    -0.07
    è¿ĩåī©
    -0.07
    POSITIVE LOGITS
    çªģåıij
    0.16
    çªģåıijäºĭ件
    0.14
    åģ¶
    0.14
    æĦıæĥ³ä¸įåĪ°
    0.13
    æĦıå¤ĸ
    0.12
    unexpected
    0.12
    åºĶæĢ¥
    0.12
    çªģå¦Ĥåħ¶
    0.12
     unexpected
    0.12
     xảy
    0.12
    Activations Density 0.085%

    No Known Activations