© 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. 7100
    Prev
    Next
    INDEX
    Explanations

    tools and instruments

    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
    ä¸įçĦ¶
    -0.10
    ä¹İ
    -0.10
    è´·
    -0.10
    acre
    -0.09
    orris
    -0.09
    ubber
    -0.08
     оÑĤдел
    -0.08
    ÑģÑĤвÑĥ
    -0.08
    apel
    -0.08
    ä¸ĭæīĭ
    -0.08
    POSITIVE LOGITS
    å¼ıçļĦ
    0.10
    ities
    0.10
     COPYING
    0.09
     baÅŁÄ±na
    0.09
    è¿ĻéĩĮçļĦ
    0.09
     McCl
    0.09
    å¼ı
    0.09
    è¿ĻæĶ¯
    0.08
    éĻĩ
    0.08
    åıĶåıĶ
    0.08
    Activations Density 0.018%

    No Known Activations