Update README.md

README.md

@@ -29,7 +29,7 @@ size_categories:
 
 It achieves **superior performance** across benchmarks like hellaswag and piqa while also achieving better loss compared to pro.
 
-**3M model trained on DCLM Pro vs Ultra, same architecture, same training budget.**
+**3M model trained on only DCLM Pro vs only Ultra, same architecture, same training budget.**
 
 | Metric | DCLM Pro | DCLM Ultra |
 |---|---:|---:|
@@ -47,10 +47,11 @@ It achieves **superior performance** across benchmarks like hellaswag and piqa w
 
 ### Dataset Sources
 
-- **Repository:** [More Information Needed]
-- **Paper:** [More Information Needed]
+- **Source Repo:** [More Information Needed]
+- **Filtered Repo:** [More Information Needed]
 
 ## Uses
 
-- Pre
-
+- PreTraining SLMs on clean, heavily filtered data.
+- PostTraining LLMs on clean, heavily filtered data.
+- PreTraining data quality benefits on SLMs research.