© 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. 11522
    Prev
    Next
    INDEX
    Explanations

    dogs, hot dogs, frankfurters

    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
    åıĮåĪĽ
    -0.09
    LD
    -0.08
    amble
    -0.08
    ible
    -0.08
    ä¸įåı¯ç¼ºå°ij
    -0.08
     Cary
    -0.08
    æ¿Ł
    -0.08
    å¤ļä¸ªåĽ½å®¶
    -0.08
     firma
    -0.08
    第äºĮç§į
    -0.08
    POSITIVE LOGITS
    abay
    0.11
    iÄį
    0.09
    illos
    0.09
    ç²¾ç¥ŀ
    0.09
    owitz
    0.09
    asti
    0.09
    çŀij
    0.09
    éĢ¼
    0.09
    æ¤ħ
    0.09
    astics
    0.09
    Activations Density 0.001%

    No Known Activations