Model save

README.md

@@ -43,7 +43,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 500
-- num_epochs: 1
+- num_epochs: 3
 
 ### Training results
 

runs/Sep26_17-12-25_nlu12a2/events.out.tfevents.1727338361.nlu12a2.236375.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53be2f555f8a1841de78cfa8b7e29c9315fd42f33e9b201b3224f1a1ffe7448c
-size 55369
+oid sha256:d47fd8d44e8a139629adea5212075efcb4357346ea01def704eb113e5bc832d3
+size 55723