Training in progress, epoch 1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de9d4e6af8c11cd29e69c392ec14d9364b5fc0b427d4a2da5ebbe57218e19cc2
+oid sha256:5578deba1dcc91c0b71b46aef08403e64da6a3a35df1fb9077b218d87533013d
 size 409103316

run-4/checkpoint-134/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c67fb7b9358f57bfdaef0c3589bb5904c67c8347682bd50405361ed65907e33
+oid sha256:5578deba1dcc91c0b71b46aef08403e64da6a3a35df1fb9077b218d87533013d
 size 409103316

run-4/checkpoint-134/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2ff7a2c56a2166efd74e24bd5712b8fd2fd7e799c8835c6efdc74ecaa1e64002
+oid sha256:229ed86a94f4b0d5761a6ef084d195ff75f9bfcf0ffc36de779d6024045852f8
 size 818327802

run-4/checkpoint-134/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c5a4f9e43d7415e006adfefe55fd5f201e0050139566d60db3f9915f9f0e69d
+oid sha256:e74c8a105948d48dcd15e7028bd2fa56cfb6cdd52de3679c246102a646100df2
 size 1064

run-4/checkpoint-134/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6610550880432129,
+  "best_metric": 0.632646918296814,
   "best_model_checkpoint": "BERT-WMM/run-4/checkpoint-134",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -10,26 +10,26 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7248826291079812,
-      "eval_loss": 0.6610550880432129,
-      "eval_runtime": 2.0768,
-      "eval_samples_per_second": 1025.597,
-      "eval_steps_per_second": 64.521,
+      "eval_accuracy": 0.7352112676056338,
+      "eval_loss": 0.632646918296814,
+      "eval_runtime": 1.9849,
+      "eval_samples_per_second": 1073.124,
+      "eval_steps_per_second": 67.511,
       "step": 134
     }
   ],
   "logging_steps": 500,
-  "max_steps": 804,
+  "max_steps": 536,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "total_flos": 0,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 1.1434279877264858e-05,
-    "num_train_epochs": 6,
+    "learning_rate": 2.9536715900078686e-05,
+    "num_train_epochs": 4,
     "per_device_train_batch_size": 64,
-    "weight_decay": 0.007798418229484321
+    "weight_decay": 0.0002947113945627026
   }
 }

run-4/checkpoint-134/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc8ba15acf69ccfa7cad78d9e146e493b9003672aa3b6919156015dfb1a7a16
+oid sha256:02d75b109551f68ad4f0f12814a2fb4545b6f0e4d94a8e75020f66e88d8a06ea
 size 4856

run-4/checkpoint-268/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc9eb0941b8fb2df459d3d00b865513510c96290e47db2cfc7345527f4e296c0
+oid sha256:6bb5f6d8e7fd35a1b474339dc030269c4e58beeb61e1aae7e7a6986484220326
 size 409103316

run-4/checkpoint-268/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:216c153733556ac4ce9db69c37cbcb2c1162e907d154a9eded2328d829e99e68
+oid sha256:bed4b64e811ecb30a9114ccec322170b49bef2f20d58f5dc9fdab1d26175ee05
 size 818327802

run-4/checkpoint-268/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f20e2d43cb43bcddd3be93609f4dfb28258ca14684d749f83306b1b2d248346f
+oid sha256:94a9c9791dff14b2da8a37b3531396d2d83be144283e29ef8ac77b77d84d0480
 size 1064

run-4/checkpoint-268/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6047325730323792,
+  "best_metric": 0.6025658845901489,
   "best_model_checkpoint": "BERT-WMM/run-4/checkpoint-268",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,35 +10,35 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7248826291079812,
-      "eval_loss": 0.6610550880432129,
-      "eval_runtime": 2.0768,
-      "eval_samples_per_second": 1025.597,
-      "eval_steps_per_second": 64.521,
+      "eval_accuracy": 0.7352112676056338,
+      "eval_loss": 0.632646918296814,
+      "eval_runtime": 1.9849,
+      "eval_samples_per_second": 1073.124,
+      "eval_steps_per_second": 67.511,
       "step": 134
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.755868544600939,
-      "eval_loss": 0.6047325730323792,
-      "eval_runtime": 2.0636,
-      "eval_samples_per_second": 1032.163,
-      "eval_steps_per_second": 64.934,
+      "eval_accuracy": 0.7539906103286385,
+      "eval_loss": 0.6025658845901489,
+      "eval_runtime": 2.0385,
+      "eval_samples_per_second": 1044.892,
+      "eval_steps_per_second": 65.735,
       "step": 268
     }
   ],
   "logging_steps": 500,
-  "max_steps": 804,
+  "max_steps": 536,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "total_flos": 0,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 1.1434279877264858e-05,
-    "num_train_epochs": 6,
+    "learning_rate": 2.9536715900078686e-05,
+    "num_train_epochs": 4,
     "per_device_train_batch_size": 64,
-    "weight_decay": 0.007798418229484321
+    "weight_decay": 0.0002947113945627026
   }
 }

run-4/checkpoint-268/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc8ba15acf69ccfa7cad78d9e146e493b9003672aa3b6919156015dfb1a7a16
+oid sha256:02d75b109551f68ad4f0f12814a2fb4545b6f0e4d94a8e75020f66e88d8a06ea
 size 4856

run-4/checkpoint-402/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fa184fe52664b1fd1422fb1186d2c131a277fe4340bbecf4f067a18ebfea8d4
+oid sha256:da11eae0edb6bae20c6c05602f501038d10490654ac342a1bed244719ae89542
 size 409103316

run-4/checkpoint-402/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4aad271e6d658426d0d4432898d11ec71c1de8756433f91545e7345a21139f6
+oid sha256:5f2fd1d1f2e6e05d61a164a598fe811ae24a405e889cb6e8a0825e012ca4273f
 size 818327802

run-4/checkpoint-402/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fe0f6a964ee8f879576dfe922482680498e583a2b9aa9084940f5fc6c0a74ee
+oid sha256:737a486a5e0651c5b0f4054192d6d4d34363f7e89b41ec47064702648d3f7b4d
 size 1064

run-4/checkpoint-402/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6047325730323792,
+  "best_metric": 0.6025658845901489,
   "best_model_checkpoint": "BERT-WMM/run-4/checkpoint-268",
   "epoch": 3.0,
   "eval_steps": 500,
@@ -10,44 +10,44 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7248826291079812,
-      "eval_loss": 0.6610550880432129,
-      "eval_runtime": 2.0768,
-      "eval_samples_per_second": 1025.597,
-      "eval_steps_per_second": 64.521,
+      "eval_accuracy": 0.7352112676056338,
+      "eval_loss": 0.632646918296814,
+      "eval_runtime": 1.9849,
+      "eval_samples_per_second": 1073.124,
+      "eval_steps_per_second": 67.511,
       "step": 134
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.755868544600939,
-      "eval_loss": 0.6047325730323792,
-      "eval_runtime": 2.0636,
-      "eval_samples_per_second": 1032.163,
-      "eval_steps_per_second": 64.934,
+      "eval_accuracy": 0.7539906103286385,
+      "eval_loss": 0.6025658845901489,
+      "eval_runtime": 2.0385,
+      "eval_samples_per_second": 1044.892,
+      "eval_steps_per_second": 65.735,
       "step": 268
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7568075117370892,
-      "eval_loss": 0.6309530138969421,
-      "eval_runtime": 2.6635,
-      "eval_samples_per_second": 799.702,
-      "eval_steps_per_second": 50.31,
+      "eval_accuracy": 0.760093896713615,
+      "eval_loss": 0.6695208549499512,
+      "eval_runtime": 2.0168,
+      "eval_samples_per_second": 1056.124,
+      "eval_steps_per_second": 66.442,
       "step": 402
     }
   ],
   "logging_steps": 500,
-  "max_steps": 804,
+  "max_steps": 536,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "total_flos": 0,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 1.1434279877264858e-05,
-    "num_train_epochs": 6,
+    "learning_rate": 2.9536715900078686e-05,
+    "num_train_epochs": 4,
     "per_device_train_batch_size": 64,
-    "weight_decay": 0.007798418229484321
+    "weight_decay": 0.0002947113945627026
   }
 }

run-4/checkpoint-402/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc8ba15acf69ccfa7cad78d9e146e493b9003672aa3b6919156015dfb1a7a16
+oid sha256:02d75b109551f68ad4f0f12814a2fb4545b6f0e4d94a8e75020f66e88d8a06ea
 size 4856

run-4/checkpoint-536/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e7180d1e19a898f1beb292a73f8386dde07cf28e7b65dcfb30d59b43d88e596
+oid sha256:b5fa86ea147adb533de5eb931defb14a81be092df5d218a025df2be2d814c7ee
 size 409103316

run-4/checkpoint-536/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0945796164298b8dae5947303aaaf9d12dcf5f2fc2419842a33b25cbf85d840c
+oid sha256:9ea0156549a0314406690dc1226e505cfbc3e0d8632b34987df784f9e62ce161
 size 818327802

run-4/checkpoint-536/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51915d7c9d6b8048cea791a1c7b08ecfce8e5f266f0131ad618c93aed2f4d5cf
+oid sha256:e5ee0f07b286c086c73cee7fc809e9534cbed92c08f78d38e5de402489003e2f
 size 1064

run-4/checkpoint-536/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.6047325730323792,
+  "best_metric": 0.6025658845901489,
   "best_model_checkpoint": "BERT-WMM/run-4/checkpoint-268",
   "epoch": 4.0,
   "eval_steps": 500,
@@ -10,60 +10,60 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7248826291079812,
-      "eval_loss": 0.6610550880432129,
-      "eval_runtime": 2.0768,
-      "eval_samples_per_second": 1025.597,
-      "eval_steps_per_second": 64.521,
+      "eval_accuracy": 0.7352112676056338,
+      "eval_loss": 0.632646918296814,
+      "eval_runtime": 1.9849,
+      "eval_samples_per_second": 1073.124,
+      "eval_steps_per_second": 67.511,
       "step": 134
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.755868544600939,
-      "eval_loss": 0.6047325730323792,
-      "eval_runtime": 2.0636,
-      "eval_samples_per_second": 1032.163,
-      "eval_steps_per_second": 64.934,
+      "eval_accuracy": 0.7539906103286385,
+      "eval_loss": 0.6025658845901489,
+      "eval_runtime": 2.0385,
+      "eval_samples_per_second": 1044.892,
+      "eval_steps_per_second": 65.735,
       "step": 268
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7568075117370892,
-      "eval_loss": 0.6309530138969421,
-      "eval_runtime": 2.6635,
-      "eval_samples_per_second": 799.702,
-      "eval_steps_per_second": 50.31,
+      "eval_accuracy": 0.760093896713615,
+      "eval_loss": 0.6695208549499512,
+      "eval_runtime": 2.0168,
+      "eval_samples_per_second": 1056.124,
+      "eval_steps_per_second": 66.442,
       "step": 402
     },
     {
       "epoch": 3.73,
-      "grad_norm": 9.669575691223145,
-      "learning_rate": 4.323409306826513e-06,
-      "loss": 0.5385,
+      "grad_norm": 11.244783401489258,
+      "learning_rate": 1.9838092768709567e-06,
+      "loss": 0.4285,
       "step": 500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7577464788732394,
-      "eval_loss": 0.6676124334335327,
-      "eval_runtime": 2.0995,
-      "eval_samples_per_second": 1014.528,
-      "eval_steps_per_second": 63.825,
+      "eval_accuracy": 0.7666666666666667,
+      "eval_loss": 0.7637132406234741,
+      "eval_runtime": 2.0114,
+      "eval_samples_per_second": 1058.984,
+      "eval_steps_per_second": 66.622,
       "step": 536
     }
   ],
   "logging_steps": 500,
-  "max_steps": 804,
+  "max_steps": 536,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 4,
   "save_steps": 500,
   "total_flos": 709335274032504.0,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 1.1434279877264858e-05,
-    "num_train_epochs": 6,
+    "learning_rate": 2.9536715900078686e-05,
+    "num_train_epochs": 4,
     "per_device_train_batch_size": 64,
-    "weight_decay": 0.007798418229484321
+    "weight_decay": 0.0002947113945627026
   }
 }

run-4/checkpoint-536/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edc8ba15acf69ccfa7cad78d9e146e493b9003672aa3b6919156015dfb1a7a16
+oid sha256:02d75b109551f68ad4f0f12814a2fb4545b6f0e4d94a8e75020f66e88d8a06ea
 size 4856

runs/Apr18_09-56-56_da5e8340a418/events.out.tfevents.1713436018.da5e8340a418.15241.24

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cff5eed09f5b19ba55a299f7f4a2f3f384ee57ac55d8d8c9179d1b02e9606fd
+size 6768

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:064739bef7fc58b5c8718d2766687070f99f1054b015389081d170a0b0689377
+oid sha256:02d75b109551f68ad4f0f12814a2fb4545b6f0e4d94a8e75020f66e88d8a06ea
 size 4856