© 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. 19633
    Prev
    Next
    INDEX
    Explanations

    different

    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
    ä¸įåĪĨ
    -0.10
    DirectoryName
    -0.10
    enia
    -0.09
    æ®ī
    -0.09
    é¢ĦåħĪ
    -0.09
    ela
    -0.08
    缸åħ³æĸ°éĹ»
    -0.08
    çĪµ
    -0.08
    å¯Łçľĭ
    -0.08
    alen
    -0.08
    POSITIVE LOGITS
    ä¸įåIJĮçļĦ
    0.19
    æľīä¸įåIJĮçļĦ
    0.18
    çļĦä¸įåIJĮ
    0.16
    个æĢ§
    0.14
    ä¸įåIJĮ
    0.14
     different
    0.13
    åı£åij³
    0.13
    Different
    0.13
    çķ°ãģªãĤĭ
    0.13
    éģķãģĨ
    0.13
    Activations Density 0.357%

    No Known Activations