© 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. 5122
    Prev
    Next
    INDEX
    Explanations

    Radcliffe and tool suffixes

    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
    è¾ĭ
    -0.10
    åı¯æĥ³
    -0.09
    ricks
    -0.09
    æ½¢
    -0.09
    å¡ĺ
    -0.08
    éĤ£äºº
    -0.08
    ernal
    -0.08
    rÃŃ
    -0.08
    éo
    -0.08
    çĸ«
    -0.08
    POSITIVE LOGITS
    鼷æĸ¯
    0.11
    presso
    0.10
    epam
    0.09
    ãĥ¼ãĥł
    0.09
    头æĻķ
    0.09
    ãĥ¼ãĥĪ
    0.09
    å¼Ģ端
    0.09
    롯
    0.09
    æŃ£æĸĩ
    0.08
    觥
    0.08
    Activations Density 0.099%

    No Known Activations