AbstractPhil commited on
Commit
5f93ebf
·
verified ·
1 Parent(s): f5e23e8

Update model card (step 78750)

Browse files
Files changed (1) hide show
  1. README.md +5 -5
README.md CHANGED
@@ -22,19 +22,19 @@ Fuses CLIP-L, CLIP-G, and decoupled T5-XL scales into a unified latent space.
22
 
23
  - **Fusion Strategy**: adaptive_cantor
24
  - **Latent Dimension**: 2048
25
- - **Training Steps**: 77,500
26
- - **Best Loss**: 0.2344
27
 
28
  ## Learned Parameters
29
 
30
  **Alpha (Visibility):**
31
  - clip_g: 0.7291
32
- - clip_l: 0.7279
33
  - t5_xl_g: 0.7244
34
- - t5_xl_l: 0.7162
35
 
36
  **Beta (Capacity):**
37
- - clip_l_t5_xl_l: 0.5720
38
  - clip_g_t5_xl_g: 0.5744
39
 
40
 
 
22
 
23
  - **Fusion Strategy**: adaptive_cantor
24
  - **Latent Dimension**: 2048
25
+ - **Training Steps**: 78,750
26
+ - **Best Loss**: 0.2336
27
 
28
  ## Learned Parameters
29
 
30
  **Alpha (Visibility):**
31
  - clip_g: 0.7291
32
+ - clip_l: 0.7280
33
  - t5_xl_g: 0.7244
34
+ - t5_xl_l: 0.7161
35
 
36
  **Beta (Capacity):**
37
+ - clip_l_t5_xl_l: 0.5726
38
  - clip_g_t5_xl_g: 0.5744
39
 
40