Update README.md

README.md

@@ -8,9 +8,16 @@ A set of embedding model trained for study embedding quality vs model architectu
 
 - **cat-emb-2-128**: 2 layers/hidden size 128/4.4m
 - **cat-emb-4-128**: 4 layers/H 128/4.8m
-- **cat-emb-6-128**: 6 layers/H 128/5.2m
 - **cat-emb-8-128**: 8 layers/H 128/5.6m
-- **cat-emb-10-128**: 10 layers/H 128/6.0m
 - **cat-emb-12-128**: 12 layers/H 128/6.4m
 - **cat-emb-2-256**: 2 layers/H 256/9.7m
-- **cat-emb-4-256**: 4 layers/H 256/11.3m
+- **cat-emb-4-256**: 4 layers/H 256/11.3m
+
+### Training
+
+- stage 1: seq 192, batch size 2048, 50k steps, sentence pairs.
+- stage 2: seq 512, batch size 64, 5k steps, sentence triplets.
+
+### Perf
+
+