© 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. 61232
    Prev
    Next
    INDEX
    Explanations

    instead of

    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
    ä¾Ľ
    -0.09
    defs
    -0.09
     bi
    -0.09
    ndl
    -0.08
    è¿ĺ
    -0.08
    ä¸Ģ度
    -0.08
    çļĨ
    -0.08
    finally
    -0.08
    omb
    -0.08
    inate
    -0.08
    POSITIVE LOGITS
    èĢĮä¸įæĺ¯
    0.26
    èĢĮéĿŀ
    0.23
     rather
    0.22
    rather
    0.22
     Rather
    0.18
    Rather
    0.17
    ãģ§ãģ¯ãģªãģı
    0.17
     chứ
    0.15
     plutôt
    0.15
    instead
    0.14
    Activations Density 0.035%

    No Known Activations