Clean README after removing partial epoch uploads

README.md

@@ -32,7 +32,7 @@ Each model is trained with multiple variants:
 | `granite-4.0-h-tiny` | base_attn, dp3_attn, dp8_attn | `ibm-granite/granite-4.0-h-tiny-base` |
 | `qwen1.5-moe-a2.7b` | dp3_attn, base_attn_v2, dp3_attn_v2, dp8_attn_v2 | `Qwen/Qwen1.5-MoE-A2.7B` |
 
-Total: **24 LoRA adapters** (final checkpoints)
+Total: **24 LoRA adapters**
 
 ## Training Data
 
@@ -57,12 +57,6 @@ Each variant directory contains:
 └── audit_scores.npz      # DP audit raw scores
 ```
 
-**Note**: Per-epoch adapter checkpoints (`epochs/epoch_XXX/adapter/`) are only partially uploaded for:
-- `llama3.2-3b/dp3` (epochs 1, 2, 3, 5)
-- `llama3.2-3b/dp8` (epochs 1, 5)
-
-Other variants only contain the final adapter. Intermediate epoch checkpoints can be requested.
-
 ## Loading a Model
 
 ```python