© 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. 1567
    Prev
    Next
    INDEX
    Explanations

    container shapes

    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
    ç§Ĩ
    -0.09
    ÑĢей
    -0.09
    褥
    -0.09
    å±ij
    -0.08
     hookers
    -0.08
    IfExists
    -0.08
     decid
    -0.08
    ruz
    -0.08
    çĶŁæĢķ
    -0.08
    顾
    -0.08
    POSITIVE LOGITS
    壳
    0.13
    çł´è£Ĥ
    0.11
    èĮ§
    0.11
    shell
    0.11
    /packages
    0.10
    å¤ĸ壳
    0.10
    hips
    0.10
     shell
    0.10
    çĤ¸å¼¹
    0.10
    .shell
    0.10
    Activations Density 0.185%

    No Known Activations