© 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. 50933
    Prev
    Next
    INDEX
    Explanations

    involved in systems

    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
    对èĩªå·±çļĦ
    -0.11
    å¹´ä¹ĭ
    -0.10
    ulen
    -0.09
    ÌĢ
    -0.09
    对æĪij们
    -0.09
    ãģĹãģ£
    -0.08
    æĪijãĢħ
    -0.08
     Malk
    -0.08
    æł¼å°Ķ
    -0.08
    濯
    -0.08
    POSITIVE LOGITS
     involved
    0.30
    åıĤä¸İ
    0.19
     involvement
    0.18
    åıĥèĪĩ
    0.16
    æ¶īåıĬ
    0.16
    æ¶īæ¡Ī
    0.15
    åıĤä¸İèĢħ
    0.15
    æ¶īåıĬåĪ°
    0.14
     involve
    0.14
     involves
    0.13
    Activations Density 0.091%

    No Known Activations