Training in progress, epoch 1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fea7a822343518cc5588fbbd4f4740020da2381f0d54605937eb5bf8dca113bf
+oid sha256:1ddac4d3a3d0da455d0e501fd182e68910ab7b0a78868896b9607f09a4a31cd3
 size 437958648

run-0/checkpoint-117/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_global_step": 78,
+  "best_global_step": 117,
   "best_metric": 0.6931407942238267,
-  "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-78",
+  "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-117",
   "epoch": 3.0,
   "eval_steps": 500,
   "global_step": 117,
@@ -11,106 +11,106 @@
   "log_history": [
     {
       "epoch": 0.2564102564102564,
-      "grad_norm": 1.662626028060913,
+      "grad_norm": 1.662625789642334,
       "learning_rate": 9.487179487179487e-05,
       "loss": 0.696,
       "step": 10
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 2.0300467014312744,
+      "grad_norm": 2.0300142765045166,
       "learning_rate": 8.974358974358975e-05,
       "loss": 0.6793,
       "step": 20
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 4.490738868713379,
+      "grad_norm": 4.4956440925598145,
       "learning_rate": 8.461538461538461e-05,
       "loss": 0.6499,
       "step": 30
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.631768953068592,
-      "eval_loss": 0.6309793591499329,
-      "eval_runtime": 0.6606,
-      "eval_samples_per_second": 419.298,
-      "eval_steps_per_second": 7.569,
+      "eval_accuracy": 0.6173285198555957,
+      "eval_loss": 0.6330550909042358,
+      "eval_runtime": 0.6611,
+      "eval_samples_per_second": 418.98,
+      "eval_steps_per_second": 7.563,
       "step": 39
     },
     {
       "epoch": 1.0256410256410255,
-      "grad_norm": 3.481339454650879,
+      "grad_norm": 3.44403338432312,
       "learning_rate": 7.948717948717948e-05,
-      "loss": 0.6793,
+      "loss": 0.6782,
       "step": 40
     },
     {
       "epoch": 1.282051282051282,
-      "grad_norm": 4.932971477508545,
+      "grad_norm": 5.425619602203369,
       "learning_rate": 7.435897435897436e-05,
-      "loss": 0.5602,
+      "loss": 0.5831,
       "step": 50
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 7.099682807922363,
+      "grad_norm": 11.772316932678223,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.5998,
+      "loss": 0.5307,
       "step": 60
     },
     {
       "epoch": 1.7948717948717947,
-      "grad_norm": 3.696152925491333,
+      "grad_norm": 3.624584674835205,
       "learning_rate": 6.410256410256412e-05,
-      "loss": 0.5403,
+      "loss": 0.5396,
       "step": 70
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6931407942238267,
-      "eval_loss": 0.6287456154823303,
-      "eval_runtime": 0.6619,
-      "eval_samples_per_second": 418.511,
-      "eval_steps_per_second": 7.554,
+      "eval_accuracy": 0.6859205776173285,
+      "eval_loss": 0.6415449380874634,
+      "eval_runtime": 0.6613,
+      "eval_samples_per_second": 418.848,
+      "eval_steps_per_second": 7.56,
       "step": 78
     },
     {
       "epoch": 2.051282051282051,
-      "grad_norm": 3.8591785430908203,
+      "grad_norm": 3.940481662750244,
       "learning_rate": 5.897435897435898e-05,
-      "loss": 0.4872,
+      "loss": 0.455,
       "step": 80
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 4.27885627746582,
+      "grad_norm": 5.196528434753418,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.3396,
+      "loss": 0.3532,
       "step": 90
     },
     {
       "epoch": 2.564102564102564,
-      "grad_norm": 4.205892562866211,
+      "grad_norm": 4.290834903717041,
       "learning_rate": 4.871794871794872e-05,
-      "loss": 0.3372,
+      "loss": 0.3482,
       "step": 100
     },
     {
       "epoch": 2.8205128205128203,
-      "grad_norm": 5.800762176513672,
+      "grad_norm": 4.902711391448975,
       "learning_rate": 4.358974358974359e-05,
-      "loss": 0.3422,
+      "loss": 0.3223,
       "step": 110
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6787003610108303,
-      "eval_loss": 0.7227063179016113,
-      "eval_runtime": 0.6696,
-      "eval_samples_per_second": 413.706,
-      "eval_steps_per_second": 7.468,
+      "eval_accuracy": 0.6931407942238267,
+      "eval_loss": 0.7661525011062622,
+      "eval_runtime": 0.6709,
+      "eval_samples_per_second": 412.863,
+      "eval_steps_per_second": 7.452,
       "step": 117
     }
   ],

run-0/checkpoint-39/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86da292e44eb636b62e95367ef4936d89b7b6baee6fe9119a8b62c9e977f0c9d
+oid sha256:1ddac4d3a3d0da455d0e501fd182e68910ab7b0a78868896b9607f09a4a31cd3
 size 437958648

run-0/checkpoint-39/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e214d4bf994f70027fe98b044d22190bb61c8b1144a91ffbfb465e093a5c2438
+oid sha256:00aaa233a5ab792d448bf134427c353ee79cad1e66b0f69aaec190e7f1ef98af
 size 876038394

run-0/checkpoint-39/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 39,
-  "best_metric": 0.631768953068592,
+  "best_metric": 0.6173285198555957,
   "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-39",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -11,32 +11,32 @@
   "log_history": [
     {
       "epoch": 0.2564102564102564,
-      "grad_norm": 1.662626028060913,
+      "grad_norm": 1.662625789642334,
       "learning_rate": 9.487179487179487e-05,
       "loss": 0.696,
       "step": 10
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 2.0300467014312744,
+      "grad_norm": 2.0300142765045166,
       "learning_rate": 8.974358974358975e-05,
       "loss": 0.6793,
       "step": 20
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 4.490738868713379,
+      "grad_norm": 4.4956440925598145,
       "learning_rate": 8.461538461538461e-05,
       "loss": 0.6499,
       "step": 30
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.631768953068592,
-      "eval_loss": 0.6309793591499329,
-      "eval_runtime": 0.6606,
-      "eval_samples_per_second": 419.298,
-      "eval_steps_per_second": 7.569,
+      "eval_accuracy": 0.6173285198555957,
+      "eval_loss": 0.6330550909042358,
+      "eval_runtime": 0.6611,
+      "eval_samples_per_second": 418.98,
+      "eval_steps_per_second": 7.563,
       "step": 39
     }
   ],

run-0/checkpoint-39/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9df7fcee919248151322838e3ee2bda70121eb4c651bc0be349c4f5f62deeaf
+oid sha256:8dedec8cb35bb4edec4b57e62ce9549f5bcd5baa05a7b09482e7f936fba95bd7
 size 5432

run-0/checkpoint-78/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee1e01f564ab20f9119e0efa77357cde7af5a4cb8d40e9d5be30ce84b313f8cb
+oid sha256:8a6bd0bd4447c86fa907715bb20e5259062d2eb35ad8c4c9ae594e0c25acd0c2
 size 876038394

run-0/checkpoint-78/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 78,
-  "best_metric": 0.6931407942238267,
+  "best_metric": 0.6859205776173285,
   "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-78",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -11,69 +11,69 @@
   "log_history": [
     {
       "epoch": 0.2564102564102564,
-      "grad_norm": 1.662626028060913,
+      "grad_norm": 1.662625789642334,
       "learning_rate": 9.487179487179487e-05,
       "loss": 0.696,
       "step": 10
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 2.0300467014312744,
+      "grad_norm": 2.0300142765045166,
       "learning_rate": 8.974358974358975e-05,
       "loss": 0.6793,
       "step": 20
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 4.490738868713379,
+      "grad_norm": 4.4956440925598145,
       "learning_rate": 8.461538461538461e-05,
       "loss": 0.6499,
       "step": 30
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.631768953068592,
-      "eval_loss": 0.6309793591499329,
-      "eval_runtime": 0.6606,
-      "eval_samples_per_second": 419.298,
-      "eval_steps_per_second": 7.569,
+      "eval_accuracy": 0.6173285198555957,
+      "eval_loss": 0.6330550909042358,
+      "eval_runtime": 0.6611,
+      "eval_samples_per_second": 418.98,
+      "eval_steps_per_second": 7.563,
       "step": 39
     },
     {
       "epoch": 1.0256410256410255,
-      "grad_norm": 3.481339454650879,
+      "grad_norm": 3.44403338432312,
       "learning_rate": 7.948717948717948e-05,
-      "loss": 0.6793,
+      "loss": 0.6782,
       "step": 40
     },
     {
       "epoch": 1.282051282051282,
-      "grad_norm": 4.932971477508545,
+      "grad_norm": 5.425619602203369,
       "learning_rate": 7.435897435897436e-05,
-      "loss": 0.5602,
+      "loss": 0.5831,
       "step": 50
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 7.099682807922363,
+      "grad_norm": 11.772316932678223,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.5998,
+      "loss": 0.5307,
       "step": 60
     },
     {
       "epoch": 1.7948717948717947,
-      "grad_norm": 3.696152925491333,
+      "grad_norm": 3.624584674835205,
       "learning_rate": 6.410256410256412e-05,
-      "loss": 0.5403,
+      "loss": 0.5396,
       "step": 70
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6931407942238267,
-      "eval_loss": 0.6287456154823303,
-      "eval_runtime": 0.6619,
-      "eval_samples_per_second": 418.511,
-      "eval_steps_per_second": 7.554,
+      "eval_accuracy": 0.6859205776173285,
+      "eval_loss": 0.6415449380874634,
+      "eval_runtime": 0.6613,
+      "eval_samples_per_second": 418.848,
+      "eval_steps_per_second": 7.56,
       "step": 78
     }
   ],

run-0/checkpoint-78/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9df7fcee919248151322838e3ee2bda70121eb4c651bc0be349c4f5f62deeaf
+oid sha256:8dedec8cb35bb4edec4b57e62ce9549f5bcd5baa05a7b09482e7f936fba95bd7
 size 5432