© 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. 36412
    Prev
    Next
    INDEX
    Explanations

    styling

    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
    æŶæĹ¶å½©
    -0.11
    bare
    -0.10
    ç½ijæĬķ
    -0.10
    halt
    -0.09
     Bare
    -0.09
    ocha
    -0.09
    峪
    -0.09
    æĸ°é²ľ
    -0.09
    æł·çļĦ
    -0.09
    odox
    -0.09
    POSITIVE LOGITS
    ria
    0.09
    ic
    0.09
    laÅŁ
    0.08
    arily
    0.08
    åĽŀäºĭ
    0.08
    ically
    0.08
    Ùij
    0.08
    istros
    0.08
    orious
    0.08
    ric
    0.08
    Activations Density 0.006%

    No Known Activations