© 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. 52551
    Prev
    Next
    INDEX
    Explanations

    diagnosed with

    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
    ynes
    -0.11
    èŁĨ
    -0.10
    ston
    -0.09
    çĿĢä»ĸ
    -0.09
     Provincial
    -0.08
    letic
    -0.08
    desired
    -0.08
    çĩ§
    -0.08
    ä¸įåģļ
    -0.08
    stein
    -0.08
    POSITIVE LOGITS
    æĹ©åľ¨
    0.09
    ement
    0.08
    赫
    0.08
    äºĨèĩªå·±çļĦ
    0.08
    åĬ²
    0.08
    åºĶ该å¦Ĥä½ķ
    0.08
    ãģ£ãģ½
    0.08
    ä»İåݻ年
    0.08
    ige
    0.08
    CDF
    0.08
    Activations Density 0.030%

    No Known Activations