Training in progress, epoch 1

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.7472924187725631,
-    "eval_loss": 0.5988962054252625,
-    "eval_runtime": 0.3009,
+    "eval_accuracy": 0.6750902527075813,
+    "eval_loss": 0.6114864945411682,
+    "eval_runtime": 0.2285,
     "eval_samples": 277,
-    "eval_samples_per_second": 920.67,
-    "eval_steps_per_second": 9.971,
-    "train_loss": 0.49518577380058093,
-    "train_runtime": 60.2408,
+    "eval_samples_per_second": 1212.012,
+    "eval_steps_per_second": 13.126,
+    "train_loss": 0.5984805376101763,
+    "train_runtime": 94.4914,
     "train_samples": 2490,
-    "train_samples_per_second": 413.341,
-    "train_steps_per_second": 12.948
+    "train_samples_per_second": 263.516,
+    "train_steps_per_second": 8.255
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.7472924187725631,
-    "eval_loss": 0.5988962054252625,
-    "eval_runtime": 0.3009,
+    "eval_accuracy": 0.6750902527075813,
+    "eval_loss": 0.6114864945411682,
+    "eval_runtime": 0.2285,
     "eval_samples": 277,
-    "eval_samples_per_second": 920.67,
-    "eval_steps_per_second": 9.971
+    "eval_samples_per_second": 1212.012,
+    "eval_steps_per_second": 13.126
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4168f966e8b785553498a9065d7db43b21cb2e0ade49d284869e1cb05acc60c2
-size 2661529
+oid sha256:479f47801246dd806b3e3b3630da4ccc2ce96a88c03ab6269bda37d649febb33
+size 2587045

runs/Jan28_14-43-03_node4/events.out.tfevents.1643352217.node4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d5adc47703bb4895f9fc48e854ef9cfc954d918c121e7e00d073c747c502733
-size 5125
+oid sha256:2b6fdbca0254bf4058ea90b07ad2d83444163328c5f706c2c19282dace102361
+size 7251

runs/Jan28_14-43-03_node4/events.out.tfevents.1643352312.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45625e5c740369a3268dfcd6176a35f4bb21697cbf945acd2e8fed7f5e10bd1d
+size 363

runs/Jan29_11-35-56_node4/1643427381.1388886/events.out.tfevents.1643427381.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c28746d03b06ee0d08bd5afb6d3210df1408bccde6037c252bd139cd9f84deee
+size 4585

runs/Jan29_11-35-56_node4/events.out.tfevents.1643427381.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba6e39695b004c5d296fc084eb130fb4c5d7c094f36dda70fc64521443f9cc43
+size 3516

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 0.49518577380058093,
-    "train_runtime": 60.2408,
+    "train_loss": 0.5984805376101763,
+    "train_runtime": 94.4914,
     "train_samples": 2490,
-    "train_samples_per_second": 413.341,
-    "train_steps_per_second": 12.948
+    "train_samples_per_second": 263.516,
+    "train_steps_per_second": 8.255
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.7472924187725631,
-  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-702",
+  "best_metric": 0.6750902527075813,
+  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-624",
   "epoch": 10.0,
   "global_step": 780,
   "is_hyper_param_search": false,
@@ -10,112 +10,112 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.700201690196991,
-      "eval_runtime": 0.2542,
-      "eval_samples_per_second": 1089.901,
-      "eval_steps_per_second": 11.804,
+      "eval_loss": 0.70159912109375,
+      "eval_runtime": 0.2776,
+      "eval_samples_per_second": 997.786,
+      "eval_steps_per_second": 10.806,
       "step": 78
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6389891696750902,
-      "eval_loss": 0.6369197964668274,
-      "eval_runtime": 0.3571,
-      "eval_samples_per_second": 775.771,
-      "eval_steps_per_second": 8.402,
+      "eval_accuracy": 0.5018050541516246,
+      "eval_loss": 0.6921780109405518,
+      "eval_runtime": 0.2113,
+      "eval_samples_per_second": 1310.929,
+      "eval_steps_per_second": 14.198,
       "step": 156
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6606498194945848,
-      "eval_loss": 0.6243730783462524,
-      "eval_runtime": 0.2436,
-      "eval_samples_per_second": 1137.183,
-      "eval_steps_per_second": 12.316,
+      "eval_accuracy": 0.5667870036101083,
+      "eval_loss": 0.6696296334266663,
+      "eval_runtime": 0.2287,
+      "eval_samples_per_second": 1211.094,
+      "eval_steps_per_second": 13.117,
       "step": 234
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7184115523465704,
-      "eval_loss": 0.572300136089325,
-      "eval_runtime": 0.228,
-      "eval_samples_per_second": 1214.712,
-      "eval_steps_per_second": 13.156,
+      "eval_accuracy": 0.631768953068592,
+      "eval_loss": 0.643951416015625,
+      "eval_runtime": 0.2582,
+      "eval_samples_per_second": 1072.841,
+      "eval_steps_per_second": 11.619,
       "step": 312
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.7220216606498195,
-      "eval_loss": 0.5677592754364014,
-      "eval_runtime": 0.2299,
-      "eval_samples_per_second": 1204.73,
-      "eval_steps_per_second": 13.048,
+      "eval_accuracy": 0.631768953068592,
+      "eval_loss": 0.6657952070236206,
+      "eval_runtime": 0.2373,
+      "eval_samples_per_second": 1167.361,
+      "eval_steps_per_second": 12.643,
       "step": 390
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.7220216606498195,
-      "eval_loss": 0.6392128467559814,
-      "eval_runtime": 0.2306,
-      "eval_samples_per_second": 1201.256,
-      "eval_steps_per_second": 13.01,
+      "eval_accuracy": 0.6570397111913358,
+      "eval_loss": 0.6278102993965149,
+      "eval_runtime": 0.2222,
+      "eval_samples_per_second": 1246.371,
+      "eval_steps_per_second": 13.499,
       "step": 468
     },
     {
       "epoch": 6.41,
       "learning_rate": 0.00019099590723055935,
-      "loss": 0.5649,
+      "loss": 0.6432,
       "step": 500
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.7364620938628159,
-      "eval_loss": 0.5661243200302124,
-      "eval_runtime": 0.3267,
-      "eval_samples_per_second": 847.854,
-      "eval_steps_per_second": 9.183,
+      "eval_accuracy": 0.6642599277978339,
+      "eval_loss": 0.6154282689094543,
+      "eval_runtime": 0.2258,
+      "eval_samples_per_second": 1226.508,
+      "eval_steps_per_second": 13.283,
       "step": 546
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.7328519855595668,
-      "eval_loss": 0.6633248329162598,
-      "eval_runtime": 0.228,
-      "eval_samples_per_second": 1214.709,
-      "eval_steps_per_second": 13.156,
+      "eval_accuracy": 0.6750902527075813,
+      "eval_loss": 0.6114864945411682,
+      "eval_runtime": 0.2213,
+      "eval_samples_per_second": 1251.904,
+      "eval_steps_per_second": 13.559,
       "step": 624
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.7472924187725631,
-      "eval_loss": 0.5988962054252625,
-      "eval_runtime": 0.2338,
-      "eval_samples_per_second": 1184.722,
-      "eval_steps_per_second": 12.831,
+      "eval_accuracy": 0.6534296028880866,
+      "eval_loss": 0.6326375603675842,
+      "eval_runtime": 0.2324,
+      "eval_samples_per_second": 1192.091,
+      "eval_steps_per_second": 12.911,
       "step": 702
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.7292418772563177,
-      "eval_loss": 0.6346577405929565,
-      "eval_runtime": 0.2362,
-      "eval_samples_per_second": 1172.727,
-      "eval_steps_per_second": 12.701,
+      "eval_accuracy": 0.6678700361010831,
+      "eval_loss": 0.6225477457046509,
+      "eval_runtime": 0.2107,
+      "eval_samples_per_second": 1314.871,
+      "eval_steps_per_second": 14.24,
       "step": 780
     },
     {
       "epoch": 10.0,
       "step": 780,
-      "total_flos": 1643505980774400.0,
-      "train_loss": 0.49518577380058093,
-      "train_runtime": 60.2408,
-      "train_samples_per_second": 413.341,
-      "train_steps_per_second": 12.948
+      "total_flos": 1638956800972800.0,
+      "train_loss": 0.5984805376101763,
+      "train_runtime": 94.4914,
+      "train_samples_per_second": 263.516,
+      "train_steps_per_second": 8.255
     }
   ],
   "max_steps": 780,
   "num_train_epochs": 10,
-  "total_flos": 1643505980774400.0,
+  "total_flos": 1638956800972800.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b70dcd2cea804f4ca61036bac5575f24f6125f32a118370cfd426b637ade935
+oid sha256:1b87fc109746fef165a5437bcaba71950ecac52b6d309d88608d3b828a6f71f9
 size 2991