Model save

README.md

@@ -39,12 +39,15 @@ The following hyperparameters were used during training:
 - train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 32
+- distributed_type: multi-GPU
+- num_devices: 3
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 384
+- total_eval_batch_size: 24
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant
 - lr_scheduler_warmup_ratio: 0.03
-- num_epochs: 2
+- num_epochs: 3
 
 ### Training results
 

runs/Sep26_14-42-15_nlu12a2/events.out.tfevents.1727329413.nlu12a2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eee863c053c8f9feca5df3459111dbb6554202d844d4e7a3513e24699f884226
-size 6506
+oid sha256:5b20ba71174df90c5be9b4f4d9f25e9a7b20ce5329ed277d03e8696478a4974c
+size 7061