© 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. 38296
    Prev
    Next
    INDEX
    Explanations

    debugging breakpoint debugger step into

    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
    Benchmark
    -0.11
    æĨĶ
    -0.10
    евиÑĩ
    -0.10
    ycz
    -0.09
     Mont
    -0.09
     Santa
    -0.08
    娶
    -0.08
    çĭĴ
    -0.08
    å¾Ĺ罪
    -0.08
    éĢģè´§
    -0.08
    POSITIVE LOGITS
    éĢĤæŶ
    0.10
    æīĵæĸŃ
    0.09
     breakpoints
    0.09
    presso
    0.09
     debugger
    0.09
    anes
    0.09
    为éĩįçĤ¹
    0.09
    约è°Ī
    0.09
     Favorite
    0.09
    Favorite
    0.08
    Activations Density 0.225%

    No Known Activations