© 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. 31446
    Prev
    Next
    INDEX
    Explanations

    I have

    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
    éĤ£æł·
    -0.10
    éĤ£ç§į
    -0.08
    weed
    -0.08
    è¥Ł
    -0.08
     wee
    -0.08
    åģķ
    -0.08
    çijĽ
    -0.08
    duit
    -0.08
    еÑģÑĤв
    -0.08
    боÑĤ
    -0.08
    POSITIVE LOGITS
    å½»
    0.10
    åĩºç§Ł
    0.09
    've
    0.09
    ãģĦãģĭ
    0.09
    èµĸ以
    0.09
    ëįĶ
    0.09
    称ä¹ĭ为
    0.09
    aped
    0.08
    reh
    0.08
    éļıæīĭ
    0.08
    Activations Density 0.116%

    No Known Activations