© 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. 21407
    Prev
    Next
    INDEX
    Explanations

    scheme

    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
    åħħæ²Ľ
    -0.11
    ted
    -0.10
    ãģĦãģĨ
    -0.10
    ãĤĪãģĨãģª
    -0.09
     vẻ
    -0.09
    çķĮçļĦ
    -0.09
    worth
    -0.09
    iable
    -0.09
    év
    -0.09
    Ìģ
    -0.09
    POSITIVE LOGITS
    forth
    0.12
    ester
    0.11
    OfWork
    0.10
    ple
    0.10
    ingo
    0.09
    atters
    0.09
    illing
    0.09
     dope
    0.09
    atics
    0.09
    ìļ¸
    0.09
    Activations Density 0.017%

    No Known Activations