© 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. 51334
    Prev
    Next
    INDEX
    Explanations

    jump on board

    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
    æĸ¹éĴĪ
    -0.10
    禽
    -0.09
    awl
    -0.09
    references
    -0.09
    yle
    -0.09
    ä¸įè§ģ
    -0.09
    stead
    -0.09
    èĨĺ
    -0.09
     Overrides
    -0.08
    æĤĸ
    -0.08
    POSITIVE LOGITS
    åıĤä¸İ
    0.14
    æĬ¢
    0.14
    åĬłåħ¥
    0.13
     góp
    0.11
    ä»ĭåħ¥
    0.11
    åıĥèĪĩ
    0.11
    æĬ¢åįł
    0.11
    åıĤä¸İåĪ°
    0.11
    æIJ¶
    0.10
     contribute
    0.10
    Activations Density 0.217%

    No Known Activations