Training in progress, epoch 4

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86da292e44eb636b62e95367ef4936d89b7b6baee6fe9119a8b62c9e977f0c9d
+oid sha256:1543d11d39ae4983ad4b56210a391fd44e53d905d92d2536f8ce4494e3795db5
 size 437958648

run-0/checkpoint-117/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17e0c9d1a3cf96b438c626370ec0758ae280b04631d45470154b5eca1293573f
+oid sha256:8672e581da9d8df927955f18ee97b1fb7bc4b72d39bba7c191508198135749ab
 size 437958648

run-0/checkpoint-117/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02f3dffe3a080aec80d4aa45517d6cb1c8020dc49d3393ae96f05506fb56d8d1
+oid sha256:e7c35b295bba54ba1b241603995ce1811c2382076f99713945e46fceacbb9937
 size 876038394

run-0/checkpoint-117/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2aa20791cd3401b748110a053f719d6902e4d9ccc845f2f5d2ff250a3d27441
+oid sha256:a9df7fcee919248151322838e3ee2bda70121eb4c651bc0be349c4f5f62deeaf
 size 5432

run-0/checkpoint-156/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45363ce679a8dfd6a6ce8f3513e67b5693b6d30b7c4329ec9c084a47504e9ba8
+oid sha256:1543d11d39ae4983ad4b56210a391fd44e53d905d92d2536f8ce4494e3795db5
 size 437958648

run-0/checkpoint-156/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ebd079703cd72b12a422caae45df454bcdc3dda626ba153bd836afb84b1093d
+oid sha256:889253018f7f4ef7fe5b3708a9b99fd213e993b42054f01aab66786ee4ea3395
 size 876038394

run-0/checkpoint-156/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 156,
-  "best_metric": 0.7003610108303249,
+  "best_metric": 0.6967509025270758,
   "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-156",
   "epoch": 4.0,
   "eval_steps": 500,
@@ -11,21 +11,21 @@
   "log_history": [
     {
       "epoch": 0.2564102564102564,
-      "grad_norm": 1.662625789642334,
+      "grad_norm": 1.662626028060913,
       "learning_rate": 9.487179487179487e-05,
       "loss": 0.696,
       "step": 10
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 2.0300352573394775,
+      "grad_norm": 2.0300467014312744,
       "learning_rate": 8.974358974358975e-05,
       "loss": 0.6793,
       "step": 20
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 4.492157936096191,
+      "grad_norm": 4.490738868713379,
       "learning_rate": 8.461538461538461e-05,
       "loss": 0.6499,
       "step": 30
@@ -33,121 +33,121 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.631768953068592,
-      "eval_loss": 0.6312768459320068,
-      "eval_runtime": 0.6623,
-      "eval_samples_per_second": 418.266,
-      "eval_steps_per_second": 7.55,
+      "eval_loss": 0.6309793591499329,
+      "eval_runtime": 0.6606,
+      "eval_samples_per_second": 419.298,
+      "eval_steps_per_second": 7.569,
       "step": 39
     },
     {
       "epoch": 1.0256410256410255,
-      "grad_norm": 3.4885644912719727,
+      "grad_norm": 3.481339454650879,
       "learning_rate": 7.948717948717948e-05,
       "loss": 0.6793,
       "step": 40
     },
     {
       "epoch": 1.282051282051282,
-      "grad_norm": 5.2225494384765625,
+      "grad_norm": 4.932971477508545,
       "learning_rate": 7.435897435897436e-05,
-      "loss": 0.5596,
+      "loss": 0.5602,
       "step": 50
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 6.484560489654541,
+      "grad_norm": 7.099682807922363,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.5713,
+      "loss": 0.5998,
       "step": 60
     },
     {
       "epoch": 1.7948717948717947,
-      "grad_norm": 4.836739540100098,
+      "grad_norm": 3.696152925491333,
       "learning_rate": 6.410256410256412e-05,
-      "loss": 0.545,
+      "loss": 0.5403,
       "step": 70
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6714801444043321,
-      "eval_loss": 0.658456563949585,
-      "eval_runtime": 0.6622,
-      "eval_samples_per_second": 418.306,
-      "eval_steps_per_second": 7.551,
+      "eval_accuracy": 0.6931407942238267,
+      "eval_loss": 0.6287456154823303,
+      "eval_runtime": 0.6619,
+      "eval_samples_per_second": 418.511,
+      "eval_steps_per_second": 7.554,
       "step": 78
     },
     {
       "epoch": 2.051282051282051,
-      "grad_norm": 6.515610218048096,
+      "grad_norm": 3.8591785430908203,
       "learning_rate": 5.897435897435898e-05,
-      "loss": 0.4786,
+      "loss": 0.4872,
       "step": 80
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 5.974998950958252,
+      "grad_norm": 4.27885627746582,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.3373,
+      "loss": 0.3396,
       "step": 90
     },
     {
       "epoch": 2.564102564102564,
-      "grad_norm": 2.976608991622925,
+      "grad_norm": 4.205892562866211,
       "learning_rate": 4.871794871794872e-05,
-      "loss": 0.3314,
+      "loss": 0.3372,
       "step": 100
     },
     {
       "epoch": 2.8205128205128203,
-      "grad_norm": 3.50764799118042,
+      "grad_norm": 5.800762176513672,
       "learning_rate": 4.358974358974359e-05,
-      "loss": 0.3235,
+      "loss": 0.3422,
       "step": 110
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6714801444043321,
-      "eval_loss": 0.7251453399658203,
-      "eval_runtime": 0.6621,
-      "eval_samples_per_second": 418.365,
-      "eval_steps_per_second": 7.552,
+      "eval_accuracy": 0.6787003610108303,
+      "eval_loss": 0.7227063179016113,
+      "eval_runtime": 0.6696,
+      "eval_samples_per_second": 413.706,
+      "eval_steps_per_second": 7.468,
       "step": 117
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 3.907212495803833,
+      "grad_norm": 3.5253195762634277,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.2728,
+      "loss": 0.2668,
       "step": 120
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 7.000370979309082,
+      "grad_norm": 11.2240629196167,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.1829,
+      "loss": 0.1835,
       "step": 130
     },
     {
       "epoch": 3.58974358974359,
-      "grad_norm": 7.436763763427734,
+      "grad_norm": 5.6398420333862305,
       "learning_rate": 2.8205128205128207e-05,
-      "loss": 0.1877,
+      "loss": 0.1725,
       "step": 140
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 7.767152786254883,
+      "grad_norm": 6.835488319396973,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.1335,
+      "loss": 0.1297,
       "step": 150
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7003610108303249,
-      "eval_loss": 0.9089646935462952,
-      "eval_runtime": 0.6606,
-      "eval_samples_per_second": 419.294,
-      "eval_steps_per_second": 7.568,
+      "eval_accuracy": 0.6967509025270758,
+      "eval_loss": 0.9852063655853271,
+      "eval_runtime": 0.6608,
+      "eval_samples_per_second": 419.212,
+      "eval_steps_per_second": 7.567,
       "step": 156
     }
   ],

run-0/checkpoint-156/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2aa20791cd3401b748110a053f719d6902e4d9ccc845f2f5d2ff250a3d27441
+oid sha256:a9df7fcee919248151322838e3ee2bda70121eb4c651bc0be349c4f5f62deeaf
 size 5432

run-0/checkpoint-195/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_global_step": 195,
-  "best_metric": 0.7111913357400722,
-  "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-195",
+  "best_global_step": 156,
+  "best_metric": 0.6967509025270758,
+  "best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_14/run-0/checkpoint-156",
   "epoch": 5.0,
   "eval_steps": 500,
   "global_step": 195,
@@ -11,21 +11,21 @@
   "log_history": [
     {
       "epoch": 0.2564102564102564,
-      "grad_norm": 1.662625789642334,
+      "grad_norm": 1.662626028060913,
       "learning_rate": 9.487179487179487e-05,
       "loss": 0.696,
       "step": 10
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 2.0300352573394775,
+      "grad_norm": 2.0300467014312744,
       "learning_rate": 8.974358974358975e-05,
       "loss": 0.6793,
       "step": 20
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 4.492157936096191,
+      "grad_norm": 4.490738868713379,
       "learning_rate": 8.461538461538461e-05,
       "loss": 0.6499,
       "step": 30
@@ -33,158 +33,158 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.631768953068592,
-      "eval_loss": 0.6312768459320068,
-      "eval_runtime": 0.6623,
-      "eval_samples_per_second": 418.266,
-      "eval_steps_per_second": 7.55,
+      "eval_loss": 0.6309793591499329,
+      "eval_runtime": 0.6606,
+      "eval_samples_per_second": 419.298,
+      "eval_steps_per_second": 7.569,
       "step": 39
     },
     {
       "epoch": 1.0256410256410255,
-      "grad_norm": 3.4885644912719727,
+      "grad_norm": 3.481339454650879,
       "learning_rate": 7.948717948717948e-05,
       "loss": 0.6793,
       "step": 40
     },
     {
       "epoch": 1.282051282051282,
-      "grad_norm": 5.2225494384765625,
+      "grad_norm": 4.932971477508545,
       "learning_rate": 7.435897435897436e-05,
-      "loss": 0.5596,
+      "loss": 0.5602,
       "step": 50
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 6.484560489654541,
+      "grad_norm": 7.099682807922363,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.5713,
+      "loss": 0.5998,
       "step": 60
     },
     {
       "epoch": 1.7948717948717947,
-      "grad_norm": 4.836739540100098,
+      "grad_norm": 3.696152925491333,
       "learning_rate": 6.410256410256412e-05,
-      "loss": 0.545,
+      "loss": 0.5403,
       "step": 70
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6714801444043321,
-      "eval_loss": 0.658456563949585,
-      "eval_runtime": 0.6622,
-      "eval_samples_per_second": 418.306,
-      "eval_steps_per_second": 7.551,
+      "eval_accuracy": 0.6931407942238267,
+      "eval_loss": 0.6287456154823303,
+      "eval_runtime": 0.6619,
+      "eval_samples_per_second": 418.511,
+      "eval_steps_per_second": 7.554,
       "step": 78
     },
     {
       "epoch": 2.051282051282051,
-      "grad_norm": 6.515610218048096,
+      "grad_norm": 3.8591785430908203,
       "learning_rate": 5.897435897435898e-05,
-      "loss": 0.4786,
+      "loss": 0.4872,
       "step": 80
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 5.974998950958252,
+      "grad_norm": 4.27885627746582,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.3373,
+      "loss": 0.3396,
       "step": 90
     },
     {
       "epoch": 2.564102564102564,
-      "grad_norm": 2.976608991622925,
+      "grad_norm": 4.205892562866211,
       "learning_rate": 4.871794871794872e-05,
-      "loss": 0.3314,
+      "loss": 0.3372,
       "step": 100
     },
     {
       "epoch": 2.8205128205128203,
-      "grad_norm": 3.50764799118042,
+      "grad_norm": 5.800762176513672,
       "learning_rate": 4.358974358974359e-05,
-      "loss": 0.3235,
+      "loss": 0.3422,
       "step": 110
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6714801444043321,
-      "eval_loss": 0.7251453399658203,
-      "eval_runtime": 0.6621,
-      "eval_samples_per_second": 418.365,
-      "eval_steps_per_second": 7.552,
+      "eval_accuracy": 0.6787003610108303,
+      "eval_loss": 0.7227063179016113,
+      "eval_runtime": 0.6696,
+      "eval_samples_per_second": 413.706,
+      "eval_steps_per_second": 7.468,
       "step": 117
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 3.907212495803833,
+      "grad_norm": 3.5253195762634277,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.2728,
+      "loss": 0.2668,
       "step": 120
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 7.000370979309082,
+      "grad_norm": 11.2240629196167,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.1829,
+      "loss": 0.1835,
       "step": 130
     },
     {
       "epoch": 3.58974358974359,
-      "grad_norm": 7.436763763427734,
+      "grad_norm": 5.6398420333862305,
       "learning_rate": 2.8205128205128207e-05,
-      "loss": 0.1877,
+      "loss": 0.1725,
       "step": 140
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 7.767152786254883,
+      "grad_norm": 6.835488319396973,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.1335,
+      "loss": 0.1297,
       "step": 150
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7003610108303249,
-      "eval_loss": 0.9089646935462952,
-      "eval_runtime": 0.6606,
-      "eval_samples_per_second": 419.294,
-      "eval_steps_per_second": 7.568,
+      "eval_accuracy": 0.6967509025270758,
+      "eval_loss": 0.9852063655853271,
+      "eval_runtime": 0.6608,
+      "eval_samples_per_second": 419.212,
+      "eval_steps_per_second": 7.567,
       "step": 156
     },
     {
       "epoch": 4.102564102564102,
-      "grad_norm": 2.6948187351226807,
+      "grad_norm": 1.7055132389068604,
       "learning_rate": 1.794871794871795e-05,
-      "loss": 0.1229,
+      "loss": 0.1088,
       "step": 160
     },
     {
       "epoch": 4.358974358974359,
-      "grad_norm": 3.5418930053710938,
+      "grad_norm": 4.614296913146973,
       "learning_rate": 1.282051282051282e-05,
-      "loss": 0.0868,
+      "loss": 0.0803,
       "step": 170
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 6.394577980041504,
+      "grad_norm": 4.053183555603027,
       "learning_rate": 7.692307692307694e-06,
-      "loss": 0.0624,
+      "loss": 0.0625,
       "step": 180
     },
     {
       "epoch": 4.871794871794872,
-      "grad_norm": 7.906170845031738,
+      "grad_norm": 7.156663417816162,
       "learning_rate": 2.564102564102564e-06,
-      "loss": 0.0608,
+      "loss": 0.0686,
       "step": 190
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.7111913357400722,
-      "eval_loss": 1.0780714750289917,
-      "eval_runtime": 0.6628,
-      "eval_samples_per_second": 417.893,
-      "eval_steps_per_second": 7.543,
+      "eval_accuracy": 0.6859205776173285,
+      "eval_loss": 1.0840200185775757,
+      "eval_runtime": 0.6626,
+      "eval_samples_per_second": 418.066,
+      "eval_steps_per_second": 7.546,
       "step": 195
     }
   ],

run-0/checkpoint-78/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87efa456251f68adc0b2b9363c9086483d78108b5a3a35553d7869669813f8d9
+oid sha256:24b4648ded986884eb769e5736bda3f39ff6235308b7ff65c5d9cae456d76b11
 size 437958648

run-0/checkpoint-78/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca0f0afc8ff9b1bbf70aee7c27e39d28e11475bb7312389acd7e3b7b91c16532
+oid sha256:ee1e01f564ab20f9119e0efa77357cde7af5a4cb8d40e9d5be30ce84b313f8cb
 size 876038394