Training in progress, step 500

all_results.json

@@ -20,9 +20,9 @@
     "predict_samples": 1568,
     "predict_samples_per_second": 0.99,
     "predict_steps_per_second": 0.042,
-    "train_loss": 1.6920310260382014,
-    "train_runtime": 827.5296,
+    "train_loss": 0.5571378069069549,
+    "train_runtime": 308.6292,
     "train_samples": 6271,
-    "train_samples_per_second": 22.734,
-    "train_steps_per_second": 0.95
+    "train_samples_per_second": 60.957,
+    "train_steps_per_second": 2.547
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7baea5f6d4d4cb969638d4204f9ecb1ad6a1aea7a6402c373495d1eca0b71e85
+oid sha256:7d4287e1a75d90d08ad45df3bb8f31d1d2e3680120d1334780594dc984516ddb
 size 647678513

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7429bd0a6bae8c971623be70159e055d55cbbcd9b54c6fd225b3141300daf5ff
-size 2108685
+oid sha256:783c7e86aedf79dd0aae603805c9591f397ca4475620b58f9a9b78151e2cfb7c
+size 2108586

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 1.6920310260382014,
-    "train_runtime": 827.5296,
+    "train_loss": 0.5571378069069549,
+    "train_runtime": 308.6292,
     "train_samples": 6271,
-    "train_samples_per_second": 22.734,
-    "train_steps_per_second": 0.95
+    "train_samples_per_second": 60.957,
+    "train_steps_per_second": 2.547
 }

trainer_state.json

@@ -3010,1727 +3010,1727 @@
     {
       "epoch": 1.91,
       "learning_rate": 1.812977099236641e-05,
-      "loss": 1.4986,
+      "loss": 1.6102,
       "step": 501
     },
     {
       "epoch": 1.92,
       "learning_rate": 1.8066157760814252e-05,
-      "loss": 1.622,
+      "loss": 1.5535,
       "step": 502
     },
     {
       "epoch": 1.92,
       "learning_rate": 1.800254452926209e-05,
-      "loss": 1.5716,
+      "loss": 1.7148,
       "step": 503
     },
     {
       "epoch": 1.92,
       "learning_rate": 1.7938931297709923e-05,
-      "loss": 1.676,
+      "loss": 1.4462,
       "step": 504
     },
     {
       "epoch": 1.93,
       "learning_rate": 1.7875318066157763e-05,
-      "loss": 1.5049,
+      "loss": 1.5138,
       "step": 505
     },
     {
       "epoch": 1.93,
       "learning_rate": 1.78117048346056e-05,
-      "loss": 1.5857,
+      "loss": 1.5897,
       "step": 506
     },
     {
       "epoch": 1.94,
       "learning_rate": 1.7748091603053434e-05,
-      "loss": 1.5618,
+      "loss": 1.5578,
       "step": 507
     },
     {
       "epoch": 1.94,
       "learning_rate": 1.7684478371501274e-05,
-      "loss": 1.623,
+      "loss": 1.4859,
       "step": 508
     },
     {
       "epoch": 1.94,
       "learning_rate": 1.762086513994911e-05,
-      "loss": 1.4365,
+      "loss": 1.5531,
       "step": 509
     },
     {
       "epoch": 1.95,
       "learning_rate": 1.7557251908396945e-05,
-      "loss": 1.5403,
+      "loss": 1.438,
       "step": 510
     },
     {
       "epoch": 1.95,
       "learning_rate": 1.7493638676844786e-05,
-      "loss": 1.5295,
+      "loss": 1.4883,
       "step": 511
     },
     {
       "epoch": 1.95,
       "learning_rate": 1.7430025445292623e-05,
-      "loss": 1.538,
+      "loss": 1.6651,
       "step": 512
     },
     {
       "epoch": 1.96,
       "learning_rate": 1.736641221374046e-05,
-      "loss": 1.5656,
+      "loss": 1.6407,
       "step": 513
     },
     {
       "epoch": 1.96,
       "learning_rate": 1.7302798982188297e-05,
-      "loss": 1.5472,
+      "loss": 1.5907,
       "step": 514
     },
     {
       "epoch": 1.97,
       "learning_rate": 1.7239185750636134e-05,
-      "loss": 1.5966,
+      "loss": 1.4254,
       "step": 515
     },
     {
       "epoch": 1.97,
       "learning_rate": 1.717557251908397e-05,
-      "loss": 1.4952,
+      "loss": 1.5548,
       "step": 516
     },
     {
       "epoch": 1.97,
       "learning_rate": 1.7111959287531808e-05,
-      "loss": 1.6362,
+      "loss": 1.6296,
       "step": 517
     },
     {
       "epoch": 1.98,
       "learning_rate": 1.7048346055979645e-05,
-      "loss": 1.5589,
+      "loss": 1.6281,
       "step": 518
     },
     {
       "epoch": 1.98,
       "learning_rate": 1.6984732824427482e-05,
-      "loss": 1.6856,
+      "loss": 1.5837,
       "step": 519
     },
     {
       "epoch": 1.98,
       "learning_rate": 1.692111959287532e-05,
-      "loss": 1.473,
+      "loss": 1.5029,
       "step": 520
     },
     {
       "epoch": 1.99,
       "learning_rate": 1.6857506361323156e-05,
-      "loss": 1.6013,
+      "loss": 1.523,
       "step": 521
     },
     {
       "epoch": 1.99,
       "learning_rate": 1.6793893129770993e-05,
-      "loss": 1.5531,
+      "loss": 1.7224,
       "step": 522
     },
     {
       "epoch": 2.0,
       "learning_rate": 1.673027989821883e-05,
-      "loss": 1.5726,
+      "loss": 1.5858,
       "step": 523
     },
     {
       "epoch": 2.0,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 1.5935,
+      "loss": 1.5929,
       "step": 524
     },
     {
       "epoch": 2.0,
       "learning_rate": 1.6603053435114505e-05,
-      "loss": 1.4443,
+      "loss": 1.5214,
       "step": 525
     },
     {
       "epoch": 2.01,
       "learning_rate": 1.653944020356234e-05,
-      "loss": 1.479,
+      "loss": 1.5286,
       "step": 526
     },
     {
       "epoch": 2.01,
       "learning_rate": 1.647582697201018e-05,
-      "loss": 1.531,
+      "loss": 1.6211,
       "step": 527
     },
     {
       "epoch": 2.02,
       "learning_rate": 1.6412213740458016e-05,
-      "loss": 1.4709,
+      "loss": 1.4927,
       "step": 528
     },
     {
       "epoch": 2.02,
       "learning_rate": 1.6348600508905853e-05,
-      "loss": 1.4644,
+      "loss": 1.6878,
       "step": 529
     },
     {
       "epoch": 2.02,
       "learning_rate": 1.628498727735369e-05,
-      "loss": 1.4627,
+      "loss": 1.5976,
       "step": 530
     },
     {
       "epoch": 2.03,
       "learning_rate": 1.6221374045801527e-05,
-      "loss": 1.4694,
+      "loss": 1.5951,
       "step": 531
     },
     {
       "epoch": 2.03,
       "learning_rate": 1.6157760814249364e-05,
-      "loss": 1.5166,
+      "loss": 1.5708,
       "step": 532
     },
     {
       "epoch": 2.03,
       "learning_rate": 1.60941475826972e-05,
-      "loss": 1.5286,
+      "loss": 1.453,
       "step": 533
     },
     {
       "epoch": 2.04,
       "learning_rate": 1.6030534351145038e-05,
-      "loss": 1.3843,
+      "loss": 1.5377,
       "step": 534
     },
     {
       "epoch": 2.04,
       "learning_rate": 1.5966921119592875e-05,
-      "loss": 1.496,
+      "loss": 1.5596,
       "step": 535
     },
     {
       "epoch": 2.05,
       "learning_rate": 1.5903307888040712e-05,
-      "loss": 1.429,
+      "loss": 1.4487,
       "step": 536
     },
     {
       "epoch": 2.05,
       "learning_rate": 1.583969465648855e-05,
-      "loss": 1.5571,
+      "loss": 1.4551,
       "step": 537
     },
     {
       "epoch": 2.05,
       "learning_rate": 1.5776081424936386e-05,
-      "loss": 1.6211,
+      "loss": 1.4983,
       "step": 538
     },
     {
       "epoch": 2.06,
       "learning_rate": 1.5712468193384224e-05,
-      "loss": 1.4718,
+      "loss": 1.5847,
       "step": 539
     },
     {
       "epoch": 2.06,
       "learning_rate": 1.5648854961832064e-05,
-      "loss": 1.6864,
+      "loss": 1.4896,
       "step": 540
     },
     {
       "epoch": 2.06,
       "learning_rate": 1.5585241730279898e-05,
-      "loss": 1.5174,
+      "loss": 1.528,
       "step": 541
     },
     {
       "epoch": 2.07,
       "learning_rate": 1.5521628498727735e-05,
-      "loss": 1.629,
+      "loss": 1.5551,
       "step": 542
     },
     {
       "epoch": 2.07,
       "learning_rate": 1.5458015267175575e-05,
-      "loss": 1.6602,
+      "loss": 1.4976,
       "step": 543
     },
     {
       "epoch": 2.08,
       "learning_rate": 1.539440203562341e-05,
-      "loss": 1.5541,
+      "loss": 1.4458,
       "step": 544
     },
     {
       "epoch": 2.08,
       "learning_rate": 1.5330788804071246e-05,
-      "loss": 1.5774,
+      "loss": 1.4596,
       "step": 545
     },
     {
       "epoch": 2.08,
       "learning_rate": 1.5267175572519086e-05,
-      "loss": 1.4806,
+      "loss": 1.5857,
       "step": 546
     },
     {
       "epoch": 2.09,
       "learning_rate": 1.5203562340966923e-05,
-      "loss": 1.4604,
+      "loss": 1.6295,
       "step": 547
     },
     {
       "epoch": 2.09,
       "learning_rate": 1.5139949109414759e-05,
-      "loss": 1.533,
+      "loss": 1.4606,
       "step": 548
     },
     {
       "epoch": 2.1,
       "learning_rate": 1.5076335877862596e-05,
-      "loss": 1.5635,
+      "loss": 1.5193,
       "step": 549
     },
     {
       "epoch": 2.1,
       "learning_rate": 1.5012722646310435e-05,
-      "loss": 1.4374,
+      "loss": 1.5228,
       "step": 550
     },
     {
       "epoch": 2.1,
       "learning_rate": 1.494910941475827e-05,
-      "loss": 1.5468,
+      "loss": 1.51,
       "step": 551
     },
     {
       "epoch": 2.11,
       "learning_rate": 1.4885496183206107e-05,
-      "loss": 1.5297,
+      "loss": 1.6053,
       "step": 552
     },
     {
       "epoch": 2.11,
       "learning_rate": 1.4821882951653946e-05,
-      "loss": 1.5429,
+      "loss": 1.5259,
       "step": 553
     },
     {
       "epoch": 2.11,
       "learning_rate": 1.4758269720101781e-05,
-      "loss": 1.5683,
+      "loss": 1.5255,
       "step": 554
     },
     {
       "epoch": 2.12,
       "learning_rate": 1.4694656488549618e-05,
-      "loss": 1.4634,
+      "loss": 1.5712,
       "step": 555
     },
     {
       "epoch": 2.12,
       "learning_rate": 1.4631043256997457e-05,
-      "loss": 1.4366,
+      "loss": 1.5506,
       "step": 556
     },
     {
       "epoch": 2.13,
       "learning_rate": 1.4567430025445294e-05,
-      "loss": 1.5887,
+      "loss": 1.6035,
       "step": 557
     },
     {
       "epoch": 2.13,
       "learning_rate": 1.450381679389313e-05,
-      "loss": 1.506,
+      "loss": 1.5887,
       "step": 558
     },
     {
       "epoch": 2.13,
       "learning_rate": 1.4440203562340968e-05,
-      "loss": 1.5557,
+      "loss": 1.6213,
       "step": 559
     },
     {
       "epoch": 2.14,
       "learning_rate": 1.4376590330788805e-05,
-      "loss": 1.3759,
+      "loss": 1.7528,
       "step": 560
     },
     {
       "epoch": 2.14,
       "learning_rate": 1.431297709923664e-05,
-      "loss": 1.5622,
+      "loss": 1.5456,
       "step": 561
     },
     {
       "epoch": 2.15,
       "learning_rate": 1.424936386768448e-05,
-      "loss": 1.6124,
+      "loss": 1.5326,
       "step": 562
     },
     {
       "epoch": 2.15,
       "learning_rate": 1.4185750636132317e-05,
-      "loss": 1.5158,
+      "loss": 1.6657,
       "step": 563
     },
     {
       "epoch": 2.15,
       "learning_rate": 1.4122137404580155e-05,
-      "loss": 1.543,
+      "loss": 1.5182,
       "step": 564
     },
     {
       "epoch": 2.16,
       "learning_rate": 1.4058524173027989e-05,
-      "loss": 1.5875,
+      "loss": 1.5033,
       "step": 565
     },
     {
       "epoch": 2.16,
       "learning_rate": 1.3994910941475828e-05,
-      "loss": 1.4717,
+      "loss": 1.5315,
       "step": 566
     },
     {
       "epoch": 2.16,
       "learning_rate": 1.3931297709923667e-05,
-      "loss": 1.532,
+      "loss": 1.4786,
       "step": 567
     },
     {
       "epoch": 2.17,
       "learning_rate": 1.38676844783715e-05,
-      "loss": 1.4993,
+      "loss": 1.5207,
       "step": 568
     },
     {
       "epoch": 2.17,
       "learning_rate": 1.3804071246819339e-05,
-      "loss": 1.494,
+      "loss": 1.5073,
       "step": 569
     },
     {
       "epoch": 2.18,
       "learning_rate": 1.3740458015267178e-05,
-      "loss": 1.4785,
+      "loss": 1.5242,
       "step": 570
     },
     {
       "epoch": 2.18,
       "learning_rate": 1.3676844783715011e-05,
-      "loss": 1.524,
+      "loss": 1.5465,
       "step": 571
     },
     {
       "epoch": 2.18,
       "learning_rate": 1.361323155216285e-05,
-      "loss": 1.4963,
+      "loss": 1.5543,
       "step": 572
     },
     {
       "epoch": 2.19,
       "learning_rate": 1.3549618320610687e-05,
-      "loss": 1.6755,
+      "loss": 1.4518,
       "step": 573
     },
     {
       "epoch": 2.19,
       "learning_rate": 1.3486005089058526e-05,
-      "loss": 1.553,
+      "loss": 1.4947,
       "step": 574
     },
     {
       "epoch": 2.19,
       "learning_rate": 1.3422391857506361e-05,
-      "loss": 1.5205,
+      "loss": 1.4802,
       "step": 575
     },
     {
       "epoch": 2.2,
       "learning_rate": 1.3358778625954198e-05,
-      "loss": 1.5081,
+      "loss": 1.5998,
       "step": 576
     },
     {
       "epoch": 2.2,
       "learning_rate": 1.3295165394402037e-05,
-      "loss": 1.5934,
+      "loss": 1.549,
       "step": 577
     },
     {
       "epoch": 2.21,
       "learning_rate": 1.3231552162849873e-05,
-      "loss": 1.6574,
+      "loss": 1.5899,
       "step": 578
     },
     {
       "epoch": 2.21,
       "learning_rate": 1.316793893129771e-05,
-      "loss": 1.5433,
+      "loss": 1.6233,
       "step": 579
     },
     {
       "epoch": 2.21,
       "learning_rate": 1.3104325699745548e-05,
-      "loss": 1.5191,
+      "loss": 1.5587,
       "step": 580
     },
     {
       "epoch": 2.22,
       "learning_rate": 1.3040712468193386e-05,
-      "loss": 1.566,
+      "loss": 1.4515,
       "step": 581
     },
     {
       "epoch": 2.22,
       "learning_rate": 1.2977099236641221e-05,
-      "loss": 1.5619,
+      "loss": 1.5081,
       "step": 582
     },
     {
       "epoch": 2.23,
       "learning_rate": 1.291348600508906e-05,
-      "loss": 1.6038,
+      "loss": 1.4917,
       "step": 583
     },
     {
       "epoch": 2.23,
       "learning_rate": 1.2849872773536897e-05,
-      "loss": 1.47,
+      "loss": 1.6199,
       "step": 584
     },
     {
       "epoch": 2.23,
       "learning_rate": 1.2786259541984732e-05,
-      "loss": 1.4102,
+      "loss": 1.4865,
       "step": 585
     },
     {
       "epoch": 2.24,
       "learning_rate": 1.2722646310432571e-05,
-      "loss": 1.6131,
+      "loss": 1.589,
       "step": 586
     },
     {
       "epoch": 2.24,
       "learning_rate": 1.2659033078880408e-05,
-      "loss": 1.4676,
+      "loss": 1.6169,
       "step": 587
     },
     {
       "epoch": 2.24,
       "learning_rate": 1.2595419847328243e-05,
-      "loss": 1.5365,
+      "loss": 1.5173,
       "step": 588
     },
     {
       "epoch": 2.25,
       "learning_rate": 1.2531806615776082e-05,
-      "loss": 1.6006,
+      "loss": 1.5147,
       "step": 589
     },
     {
       "epoch": 2.25,
       "learning_rate": 1.2468193384223919e-05,
-      "loss": 1.4513,
+      "loss": 1.5146,
       "step": 590
     },
     {
       "epoch": 2.26,
       "learning_rate": 1.2404580152671756e-05,
-      "loss": 1.5827,
+      "loss": 1.5204,
       "step": 591
     },
     {
       "epoch": 2.26,
       "learning_rate": 1.2340966921119595e-05,
-      "loss": 1.5173,
+      "loss": 1.5539,
       "step": 592
     },
     {
       "epoch": 2.26,
       "learning_rate": 1.227735368956743e-05,
-      "loss": 1.504,
+      "loss": 1.4935,
       "step": 593
     },
     {
       "epoch": 2.27,
       "learning_rate": 1.2213740458015267e-05,
-      "loss": 1.474,
+      "loss": 1.5304,
       "step": 594
     },
     {
       "epoch": 2.27,
       "learning_rate": 1.2150127226463104e-05,
-      "loss": 1.6105,
+      "loss": 1.4887,
       "step": 595
     },
     {
       "epoch": 2.27,
       "learning_rate": 1.2086513994910942e-05,
-      "loss": 1.5222,
+      "loss": 1.5163,
       "step": 596
     },
     {
       "epoch": 2.28,
       "learning_rate": 1.202290076335878e-05,
-      "loss": 1.5107,
+      "loss": 1.5368,
       "step": 597
     },
     {
       "epoch": 2.28,
       "learning_rate": 1.1959287531806616e-05,
-      "loss": 1.4677,
+      "loss": 1.6242,
       "step": 598
     },
     {
       "epoch": 2.29,
       "learning_rate": 1.1895674300254453e-05,
-      "loss": 1.4446,
+      "loss": 1.46,
       "step": 599
     },
     {
       "epoch": 2.29,
       "learning_rate": 1.1832061068702292e-05,
-      "loss": 1.5138,
+      "loss": 1.4866,
       "step": 600
     },
     {
       "epoch": 2.29,
       "learning_rate": 1.1768447837150127e-05,
-      "loss": 1.4841,
+      "loss": 1.6052,
       "step": 601
     },
     {
       "epoch": 2.3,
       "learning_rate": 1.1704834605597966e-05,
-      "loss": 1.526,
+      "loss": 1.5368,
       "step": 602
     },
     {
       "epoch": 2.3,
       "learning_rate": 1.1641221374045803e-05,
-      "loss": 1.4385,
+      "loss": 1.5202,
       "step": 603
     },
     {
       "epoch": 2.31,
       "learning_rate": 1.1577608142493638e-05,
-      "loss": 1.5106,
+      "loss": 1.4896,
       "step": 604
     },
     {
       "epoch": 2.31,
       "learning_rate": 1.1513994910941477e-05,
-      "loss": 1.4731,
+      "loss": 1.542,
       "step": 605
     },
     {
       "epoch": 2.31,
       "learning_rate": 1.1450381679389314e-05,
-      "loss": 1.4147,
+      "loss": 1.5377,
       "step": 606
     },
     {
       "epoch": 2.32,
       "learning_rate": 1.1386768447837151e-05,
-      "loss": 1.5427,
+      "loss": 1.4361,
       "step": 607
     },
     {
       "epoch": 2.32,
       "learning_rate": 1.1323155216284988e-05,
-      "loss": 1.5735,
+      "loss": 1.5759,
       "step": 608
     },
     {
       "epoch": 2.32,
       "learning_rate": 1.1259541984732823e-05,
-      "loss": 1.5151,
+      "loss": 1.488,
       "step": 609
     },
     {
       "epoch": 2.33,
       "learning_rate": 1.1195928753180662e-05,
-      "loss": 1.4713,
+      "loss": 1.604,
       "step": 610
     },
     {
       "epoch": 2.33,
       "learning_rate": 1.11323155216285e-05,
-      "loss": 1.5663,
+      "loss": 1.4868,
       "step": 611
     },
     {
       "epoch": 2.34,
       "learning_rate": 1.1068702290076336e-05,
-      "loss": 1.4052,
+      "loss": 1.6055,
       "step": 612
     },
     {
       "epoch": 2.34,
       "learning_rate": 1.1005089058524173e-05,
-      "loss": 1.5848,
+      "loss": 1.7024,
       "step": 613
     },
     {
       "epoch": 2.34,
       "learning_rate": 1.0941475826972012e-05,
-      "loss": 1.4521,
+      "loss": 1.6354,
       "step": 614
     },
     {
       "epoch": 2.35,
       "learning_rate": 1.0877862595419848e-05,
-      "loss": 1.5355,
+      "loss": 1.4439,
       "step": 615
     },
     {
       "epoch": 2.35,
       "learning_rate": 1.0814249363867685e-05,
-      "loss": 1.6279,
+      "loss": 1.5649,
       "step": 616
     },
     {
       "epoch": 2.35,
       "learning_rate": 1.0750636132315522e-05,
-      "loss": 1.4912,
+      "loss": 1.4975,
       "step": 617
     },
     {
       "epoch": 2.36,
       "learning_rate": 1.0687022900763359e-05,
-      "loss": 1.6003,
+      "loss": 1.5618,
       "step": 618
     },
     {
       "epoch": 2.36,
       "learning_rate": 1.0623409669211198e-05,
-      "loss": 1.4327,
+      "loss": 1.5837,
       "step": 619
     },
     {
       "epoch": 2.37,
       "learning_rate": 1.0559796437659033e-05,
-      "loss": 1.4725,
+      "loss": 1.6462,
       "step": 620
     },
     {
       "epoch": 2.37,
       "learning_rate": 1.049618320610687e-05,
-      "loss": 1.5927,
+      "loss": 1.576,
       "step": 621
     },
     {
       "epoch": 2.37,
       "learning_rate": 1.0432569974554709e-05,
-      "loss": 1.4698,
+      "loss": 1.5261,
       "step": 622
     },
     {
       "epoch": 2.38,
       "learning_rate": 1.0368956743002544e-05,
-      "loss": 1.568,
+      "loss": 1.5401,
       "step": 623
     },
     {
       "epoch": 2.38,
       "learning_rate": 1.0305343511450383e-05,
-      "loss": 1.5145,
+      "loss": 1.5797,
       "step": 624
     },
     {
       "epoch": 2.39,
       "learning_rate": 1.024173027989822e-05,
-      "loss": 1.479,
+      "loss": 1.5965,
       "step": 625
     },
     {
       "epoch": 2.39,
       "learning_rate": 1.0178117048346055e-05,
-      "loss": 1.5341,
+      "loss": 1.515,
       "step": 626
     },
     {
       "epoch": 2.39,
       "learning_rate": 1.0114503816793894e-05,
-      "loss": 1.455,
+      "loss": 1.54,
       "step": 627
     },
     {
       "epoch": 2.4,
       "learning_rate": 1.0050890585241731e-05,
-      "loss": 1.5176,
+      "loss": 1.5752,
       "step": 628
     },
     {
       "epoch": 2.4,
       "learning_rate": 9.987277353689568e-06,
-      "loss": 1.5831,
+      "loss": 1.5602,
       "step": 629
     },
     {
       "epoch": 2.4,
       "learning_rate": 9.923664122137405e-06,
-      "loss": 1.6207,
+      "loss": 1.5375,
       "step": 630
     },
     {
       "epoch": 2.41,
       "learning_rate": 9.860050890585242e-06,
-      "loss": 1.5222,
+      "loss": 1.63,
       "step": 631
     },
     {
       "epoch": 2.41,
       "learning_rate": 9.79643765903308e-06,
-      "loss": 1.4343,
+      "loss": 1.6018,
       "step": 632
     },
     {
       "epoch": 2.42,
       "learning_rate": 9.732824427480917e-06,
-      "loss": 1.4709,
+      "loss": 1.5341,
       "step": 633
     },
     {
       "epoch": 2.42,
       "learning_rate": 9.669211195928754e-06,
-      "loss": 1.5012,
+      "loss": 1.5728,
       "step": 634
     },
     {
       "epoch": 2.42,
       "learning_rate": 9.60559796437659e-06,
-      "loss": 1.4925,
+      "loss": 1.4774,
       "step": 635
     },
     {
       "epoch": 2.43,
       "learning_rate": 9.541984732824428e-06,
-      "loss": 1.4522,
+      "loss": 1.5961,
       "step": 636
     },
     {
       "epoch": 2.43,
       "learning_rate": 9.478371501272265e-06,
-      "loss": 1.513,
+      "loss": 1.5615,
       "step": 637
     },
     {
       "epoch": 2.44,
       "learning_rate": 9.414758269720102e-06,
-      "loss": 1.459,
+      "loss": 1.5452,
       "step": 638
     },
     {
       "epoch": 2.44,
       "learning_rate": 9.351145038167939e-06,
-      "loss": 1.4932,
+      "loss": 1.5388,
       "step": 639
     },
     {
       "epoch": 2.44,
       "learning_rate": 9.287531806615776e-06,
-      "loss": 1.4743,
+      "loss": 1.341,
       "step": 640
     },
     {
       "epoch": 2.45,
       "learning_rate": 9.223918575063615e-06,
-      "loss": 1.522,
+      "loss": 1.3478,
       "step": 641
     },
     {
       "epoch": 2.45,
       "learning_rate": 9.16030534351145e-06,
-      "loss": 1.593,
+      "loss": 1.496,
       "step": 642
     },
     {
       "epoch": 2.45,
       "learning_rate": 9.096692111959287e-06,
-      "loss": 1.6271,
+      "loss": 1.425,
       "step": 643
     },
     {
       "epoch": 2.46,
       "learning_rate": 9.033078880407126e-06,
-      "loss": 1.5129,
+      "loss": 1.6387,
       "step": 644
     },
     {
       "epoch": 2.46,
       "learning_rate": 8.969465648854961e-06,
-      "loss": 1.4341,
+      "loss": 1.5034,
       "step": 645
     },
     {
       "epoch": 2.47,
       "learning_rate": 8.9058524173028e-06,
-      "loss": 1.5421,
+      "loss": 1.7018,
       "step": 646
     },
     {
       "epoch": 2.47,
       "learning_rate": 8.842239185750637e-06,
-      "loss": 1.5236,
+      "loss": 1.384,
       "step": 647
     },
     {
       "epoch": 2.47,
       "learning_rate": 8.778625954198473e-06,
-      "loss": 1.5255,
+      "loss": 1.5595,
       "step": 648
     },
     {
       "epoch": 2.48,
       "learning_rate": 8.715012722646311e-06,
-      "loss": 1.4372,
+      "loss": 1.5429,
       "step": 649
     },
     {
       "epoch": 2.48,
       "learning_rate": 8.651399491094148e-06,
-      "loss": 1.5087,
+      "loss": 1.4169,
       "step": 650
     },
     {
       "epoch": 2.48,
       "learning_rate": 8.587786259541985e-06,
-      "loss": 1.5081,
+      "loss": 1.4469,
       "step": 651
     },
     {
       "epoch": 2.49,
       "learning_rate": 8.524173027989823e-06,
-      "loss": 1.4761,
+      "loss": 1.5503,
       "step": 652
     },
     {
       "epoch": 2.49,
       "learning_rate": 8.46055979643766e-06,
-      "loss": 1.4252,
+      "loss": 1.4019,
       "step": 653
     },
     {
       "epoch": 2.5,
       "learning_rate": 8.396946564885497e-06,
-      "loss": 1.5772,
+      "loss": 1.392,
       "step": 654
     },
     {
       "epoch": 2.5,
       "learning_rate": 8.333333333333334e-06,
-      "loss": 1.3853,
+      "loss": 1.4966,
       "step": 655
     },
     {
       "epoch": 2.5,
       "learning_rate": 8.26972010178117e-06,
-      "loss": 1.5134,
+      "loss": 1.3869,
       "step": 656
     },
     {
       "epoch": 2.51,
       "learning_rate": 8.206106870229008e-06,
-      "loss": 1.4493,
+      "loss": 1.5947,
       "step": 657
     },
     {
       "epoch": 2.51,
       "learning_rate": 8.142493638676845e-06,
-      "loss": 1.4715,
+      "loss": 1.6324,
       "step": 658
     },
     {
       "epoch": 2.52,
       "learning_rate": 8.078880407124682e-06,
-      "loss": 1.5163,
+      "loss": 1.5893,
       "step": 659
     },
     {
       "epoch": 2.52,
       "learning_rate": 8.015267175572519e-06,
-      "loss": 1.5462,
+      "loss": 1.4458,
       "step": 660
     },
     {
       "epoch": 2.52,
       "learning_rate": 7.951653944020356e-06,
-      "loss": 1.6523,
+      "loss": 1.4867,
       "step": 661
     },
     {
       "epoch": 2.53,
       "learning_rate": 7.888040712468193e-06,
-      "loss": 1.6249,
+      "loss": 1.6153,
       "step": 662
     },
     {
       "epoch": 2.53,
       "learning_rate": 7.824427480916032e-06,
-      "loss": 1.6846,
+      "loss": 1.444,
       "step": 663
     },
     {
       "epoch": 2.53,
       "learning_rate": 7.760814249363867e-06,
-      "loss": 1.6154,
+      "loss": 1.558,
       "step": 664
     },
     {
       "epoch": 2.54,
       "learning_rate": 7.697201017811704e-06,
-      "loss": 1.6876,
+      "loss": 1.5311,
       "step": 665
     },
     {
       "epoch": 2.54,
       "learning_rate": 7.633587786259543e-06,
-      "loss": 1.506,
+      "loss": 1.5768,
       "step": 666
     },
     {
       "epoch": 2.55,
       "learning_rate": 7.569974554707379e-06,
-      "loss": 1.5859,
+      "loss": 1.4543,
       "step": 667
     },
     {
       "epoch": 2.55,
       "learning_rate": 7.506361323155217e-06,
-      "loss": 1.5522,
+      "loss": 1.4829,
       "step": 668
     },
     {
       "epoch": 2.55,
       "learning_rate": 7.4427480916030536e-06,
-      "loss": 1.6109,
+      "loss": 1.4837,
       "step": 669
     },
     {
       "epoch": 2.56,
       "learning_rate": 7.379134860050891e-06,
-      "loss": 1.5327,
+      "loss": 1.4309,
       "step": 670
     },
     {
       "epoch": 2.56,
       "learning_rate": 7.3155216284987285e-06,
-      "loss": 1.4753,
+      "loss": 1.579,
       "step": 671
     },
     {
       "epoch": 2.56,
       "learning_rate": 7.251908396946565e-06,
-      "loss": 1.5325,
+      "loss": 1.3897,
       "step": 672
     },
     {
       "epoch": 2.57,
       "learning_rate": 7.188295165394403e-06,
-      "loss": 1.5384,
+      "loss": 1.4868,
       "step": 673
     },
     {
       "epoch": 2.57,
       "learning_rate": 7.12468193384224e-06,
-      "loss": 1.5745,
+      "loss": 1.5413,
       "step": 674
     },
     {
       "epoch": 2.58,
       "learning_rate": 7.061068702290078e-06,
-      "loss": 1.4398,
+      "loss": 1.5012,
       "step": 675
     },
     {
       "epoch": 2.58,
       "learning_rate": 6.997455470737914e-06,
-      "loss": 1.4848,
+      "loss": 1.4479,
       "step": 676
     },
     {
       "epoch": 2.58,
       "learning_rate": 6.93384223918575e-06,
-      "loss": 1.3778,
+      "loss": 1.6027,
       "step": 677
     },
     {
       "epoch": 2.59,
       "learning_rate": 6.870229007633589e-06,
-      "loss": 1.5155,
+      "loss": 1.5398,
       "step": 678
     },
     {
       "epoch": 2.59,
       "learning_rate": 6.806615776081425e-06,
-      "loss": 1.5289,
+      "loss": 1.4343,
       "step": 679
     },
     {
       "epoch": 2.6,
       "learning_rate": 6.743002544529263e-06,
-      "loss": 1.5638,
+      "loss": 1.4993,
       "step": 680
     },
     {
       "epoch": 2.6,
       "learning_rate": 6.679389312977099e-06,
-      "loss": 1.3992,
+      "loss": 1.5968,
       "step": 681
     },
     {
       "epoch": 2.6,
       "learning_rate": 6.615776081424936e-06,
-      "loss": 1.5364,
+      "loss": 1.573,
       "step": 682
     },
     {
       "epoch": 2.61,
       "learning_rate": 6.552162849872774e-06,
-      "loss": 1.5832,
+      "loss": 1.5111,
       "step": 683
     },
     {
       "epoch": 2.61,
       "learning_rate": 6.4885496183206104e-06,
-      "loss": 1.4741,
+      "loss": 1.5173,
       "step": 684
     },
     {
       "epoch": 2.61,
       "learning_rate": 6.424936386768448e-06,
-      "loss": 1.5595,
+      "loss": 1.4858,
       "step": 685
     },
     {
       "epoch": 2.62,
       "learning_rate": 6.3613231552162854e-06,
-      "loss": 1.5912,
+      "loss": 1.4872,
       "step": 686
     },
     {
       "epoch": 2.62,
       "learning_rate": 6.297709923664122e-06,
-      "loss": 1.5435,
+      "loss": 1.6585,
       "step": 687
     },
     {
       "epoch": 2.63,
       "learning_rate": 6.2340966921119596e-06,
-      "loss": 1.4681,
+      "loss": 1.5683,
       "step": 688
     },
     {
       "epoch": 2.63,
       "learning_rate": 6.1704834605597975e-06,
-      "loss": 1.4535,
+      "loss": 1.552,
       "step": 689
     },
     {
       "epoch": 2.63,
       "learning_rate": 6.106870229007634e-06,
-      "loss": 1.575,
+      "loss": 1.5064,
       "step": 690
     },
     {
       "epoch": 2.64,
       "learning_rate": 6.043256997455471e-06,
-      "loss": 1.4966,
+      "loss": 1.6124,
       "step": 691
     },
     {
       "epoch": 2.64,
       "learning_rate": 5.979643765903308e-06,
-      "loss": 1.5858,
+      "loss": 1.5318,
       "step": 692
     },
     {
       "epoch": 2.65,
       "learning_rate": 5.916030534351146e-06,
-      "loss": 1.4563,
+      "loss": 1.4545,
       "step": 693
     },
     {
       "epoch": 2.65,
       "learning_rate": 5.852417302798983e-06,
-      "loss": 1.4964,
+      "loss": 1.5969,
       "step": 694
     },
     {
       "epoch": 2.65,
       "learning_rate": 5.788804071246819e-06,
-      "loss": 1.5931,
+      "loss": 1.4931,
       "step": 695
     },
     {
       "epoch": 2.66,
       "learning_rate": 5.725190839694657e-06,
-      "loss": 1.3864,
+      "loss": 1.6001,
       "step": 696
     },
     {
       "epoch": 2.66,
       "learning_rate": 5.661577608142494e-06,
-      "loss": 1.422,
+      "loss": 1.5452,
       "step": 697
     },
     {
       "epoch": 2.66,
       "learning_rate": 5.597964376590331e-06,
-      "loss": 1.5293,
+      "loss": 1.4818,
       "step": 698
     },
     {
       "epoch": 2.67,
       "learning_rate": 5.534351145038168e-06,
-      "loss": 1.4404,
+      "loss": 1.5295,
       "step": 699
     },
     {
       "epoch": 2.67,
       "learning_rate": 5.470737913486006e-06,
-      "loss": 1.475,
+      "loss": 1.4386,
       "step": 700
     },
     {
       "epoch": 2.68,
       "learning_rate": 5.407124681933842e-06,
-      "loss": 1.4067,
+      "loss": 1.6439,
       "step": 701
     },
     {
       "epoch": 2.68,
       "learning_rate": 5.343511450381679e-06,
-      "loss": 1.4764,
+      "loss": 1.5542,
       "step": 702
     },
     {
       "epoch": 2.68,
       "learning_rate": 5.2798982188295165e-06,
-      "loss": 1.5613,
+      "loss": 1.5447,
       "step": 703
     },
     {
       "epoch": 2.69,
       "learning_rate": 5.216284987277354e-06,
-      "loss": 1.5702,
+      "loss": 1.5099,
       "step": 704
     },
     {
       "epoch": 2.69,
       "learning_rate": 5.1526717557251914e-06,
-      "loss": 1.4421,
+      "loss": 1.5497,
       "step": 705
     },
     {
       "epoch": 2.69,
       "learning_rate": 5.089058524173028e-06,
-      "loss": 1.5176,
+      "loss": 1.5273,
       "step": 706
     },
     {
       "epoch": 2.7,
       "learning_rate": 5.025445292620866e-06,
-      "loss": 1.5791,
+      "loss": 1.5756,
       "step": 707
     },
     {
       "epoch": 2.7,
       "learning_rate": 4.961832061068703e-06,
-      "loss": 1.6155,
+      "loss": 1.4347,
       "step": 708
     },
     {
       "epoch": 2.71,
       "learning_rate": 4.89821882951654e-06,
-      "loss": 1.3407,
+      "loss": 1.4573,
       "step": 709
     },
     {
       "epoch": 2.71,
       "learning_rate": 4.834605597964377e-06,
-      "loss": 1.5376,
+      "loss": 1.531,
       "step": 710
     },
     {
       "epoch": 2.71,
       "learning_rate": 4.770992366412214e-06,
-      "loss": 1.4459,
+      "loss": 1.5594,
       "step": 711
     },
     {
       "epoch": 2.72,
       "learning_rate": 4.707379134860051e-06,
-      "loss": 1.4462,
+      "loss": 1.4137,
       "step": 712
     },
     {
       "epoch": 2.72,
       "learning_rate": 4.643765903307888e-06,
-      "loss": 1.4415,
+      "loss": 1.4758,
       "step": 713
     },
     {
       "epoch": 2.73,
       "learning_rate": 4.580152671755725e-06,
-      "loss": 1.4233,
+      "loss": 1.5364,
       "step": 714
     },
     {
       "epoch": 2.73,
       "learning_rate": 4.516539440203563e-06,
-      "loss": 1.5232,
+      "loss": 1.5362,
       "step": 715
     },
     {
       "epoch": 2.73,
       "learning_rate": 4.4529262086514e-06,
-      "loss": 1.4517,
+      "loss": 1.5036,
       "step": 716
     },
     {
       "epoch": 2.74,
       "learning_rate": 4.389312977099236e-06,
-      "loss": 1.4326,
+      "loss": 1.4762,
       "step": 717
     },
     {
       "epoch": 2.74,
       "learning_rate": 4.325699745547074e-06,
-      "loss": 1.527,
+      "loss": 1.6156,
       "step": 718
     },
     {
       "epoch": 2.74,
       "learning_rate": 4.262086513994911e-06,
-      "loss": 1.5301,
+      "loss": 1.6199,
       "step": 719
     },
     {
       "epoch": 2.75,
       "learning_rate": 4.198473282442748e-06,
-      "loss": 1.5043,
+      "loss": 1.4238,
       "step": 720
     },
     {
       "epoch": 2.75,
       "learning_rate": 4.134860050890585e-06,
-      "loss": 1.549,
+      "loss": 1.4756,
       "step": 721
     },
     {
       "epoch": 2.76,
       "learning_rate": 4.0712468193384225e-06,
-      "loss": 1.5779,
+      "loss": 1.5169,
       "step": 722
     },
     {
       "epoch": 2.76,
       "learning_rate": 4.0076335877862595e-06,
-      "loss": 1.437,
+      "loss": 1.5376,
       "step": 723
     },
     {
       "epoch": 2.76,
       "learning_rate": 3.944020356234097e-06,
-      "loss": 1.5213,
+      "loss": 1.5023,
       "step": 724
     },
     {
       "epoch": 2.77,
       "learning_rate": 3.880407124681934e-06,
-      "loss": 1.4216,
+      "loss": 1.5147,
       "step": 725
     },
     {
       "epoch": 2.77,
       "learning_rate": 3.816793893129772e-06,
-      "loss": 1.4927,
+      "loss": 1.5219,
       "step": 726
     },
     {
       "epoch": 2.77,
       "learning_rate": 3.7531806615776087e-06,
-      "loss": 1.4148,
+      "loss": 1.4531,
       "step": 727
     },
     {
       "epoch": 2.78,
       "learning_rate": 3.6895674300254453e-06,
-      "loss": 1.5635,
+      "loss": 1.6463,
       "step": 728
     },
     {
       "epoch": 2.78,
       "learning_rate": 3.6259541984732824e-06,
-      "loss": 1.5999,
+      "loss": 1.5067,
       "step": 729
     },
     {
       "epoch": 2.79,
       "learning_rate": 3.56234096692112e-06,
-      "loss": 1.5413,
+      "loss": 1.5814,
       "step": 730
     },
     {
       "epoch": 2.79,
       "learning_rate": 3.498727735368957e-06,
-      "loss": 1.6005,
+      "loss": 1.5162,
       "step": 731
     },
     {
       "epoch": 2.79,
       "learning_rate": 3.4351145038167944e-06,
-      "loss": 1.3934,
+      "loss": 1.5473,
       "step": 732
     },
     {
       "epoch": 2.8,
       "learning_rate": 3.3715012722646315e-06,
-      "loss": 1.4354,
+      "loss": 1.5247,
       "step": 733
     },
     {
       "epoch": 2.8,
       "learning_rate": 3.307888040712468e-06,
-      "loss": 1.4843,
+      "loss": 1.6345,
       "step": 734
     },
     {
       "epoch": 2.81,
       "learning_rate": 3.2442748091603052e-06,
-      "loss": 1.4673,
+      "loss": 1.5223,
       "step": 735
     },
     {
       "epoch": 2.81,
       "learning_rate": 3.1806615776081427e-06,
-      "loss": 1.4752,
+      "loss": 1.5291,
       "step": 736
     },
     {
       "epoch": 2.81,
       "learning_rate": 3.1170483460559798e-06,
-      "loss": 1.6207,
+      "loss": 1.446,
       "step": 737
     },
     {
       "epoch": 2.82,
       "learning_rate": 3.053435114503817e-06,
-      "loss": 1.5183,
+      "loss": 1.4595,
       "step": 738
     },
     {
       "epoch": 2.82,
       "learning_rate": 2.989821882951654e-06,
-      "loss": 1.5892,
+      "loss": 1.4902,
       "step": 739
     },
     {
       "epoch": 2.82,
       "learning_rate": 2.9262086513994914e-06,
-      "loss": 1.5019,
+      "loss": 1.4615,
       "step": 740
     },
     {
       "epoch": 2.83,
       "learning_rate": 2.8625954198473285e-06,
-      "loss": 1.4875,
+      "loss": 1.4374,
       "step": 741
     },
     {
       "epoch": 2.83,
       "learning_rate": 2.7989821882951656e-06,
-      "loss": 1.4715,
+      "loss": 1.5031,
       "step": 742
     },
     {
       "epoch": 2.84,
       "learning_rate": 2.735368956743003e-06,
-      "loss": 1.4148,
+      "loss": 1.48,
       "step": 743
     },
     {
       "epoch": 2.84,
       "learning_rate": 2.6717557251908397e-06,
-      "loss": 1.5031,
+      "loss": 1.4623,
       "step": 744
     },
     {
       "epoch": 2.84,
       "learning_rate": 2.608142493638677e-06,
-      "loss": 1.5024,
+      "loss": 1.4165,
       "step": 745
     },
     {
       "epoch": 2.85,
       "learning_rate": 2.544529262086514e-06,
-      "loss": 1.4872,
+      "loss": 1.4686,
       "step": 746
     },
     {
       "epoch": 2.85,
       "learning_rate": 2.4809160305343513e-06,
-      "loss": 1.5258,
+      "loss": 1.4304,
       "step": 747
     },
     {
       "epoch": 2.85,
       "learning_rate": 2.4173027989821884e-06,
-      "loss": 1.4141,
+      "loss": 1.4525,
       "step": 748
     },
     {
       "epoch": 2.86,
       "learning_rate": 2.3536895674300255e-06,
-      "loss": 1.5452,
+      "loss": 1.4141,
       "step": 749
     },
     {
       "epoch": 2.86,
       "learning_rate": 2.2900763358778625e-06,
-      "loss": 1.4944,
+      "loss": 1.5207,
       "step": 750
     },
     {
       "epoch": 2.87,
       "learning_rate": 2.2264631043257e-06,
-      "loss": 1.4778,
+      "loss": 1.5353,
       "step": 751
     },
     {
       "epoch": 2.87,
       "learning_rate": 2.162849872773537e-06,
-      "loss": 1.4943,
+      "loss": 1.4781,
       "step": 752
     },
     {
       "epoch": 2.87,
       "learning_rate": 2.099236641221374e-06,
-      "loss": 1.5323,
+      "loss": 1.5691,
       "step": 753
     },
     {
       "epoch": 2.88,
       "learning_rate": 2.0356234096692112e-06,
-      "loss": 1.4696,
+      "loss": 1.6929,
       "step": 754
     },
     {
       "epoch": 2.88,
       "learning_rate": 1.9720101781170483e-06,
-      "loss": 1.4476,
+      "loss": 1.6577,
       "step": 755
     },
     {
       "epoch": 2.89,
       "learning_rate": 1.908396946564886e-06,
-      "loss": 1.4159,
+      "loss": 1.4755,
       "step": 756
     },
     {
       "epoch": 2.89,
       "learning_rate": 1.8447837150127227e-06,
-      "loss": 1.5014,
+      "loss": 1.6568,
       "step": 757
     },
     {
       "epoch": 2.89,
       "learning_rate": 1.78117048346056e-06,
-      "loss": 1.5421,
+      "loss": 1.4796,
       "step": 758
     },
     {
       "epoch": 2.9,
       "learning_rate": 1.7175572519083972e-06,
-      "loss": 1.499,
+      "loss": 1.5003,
       "step": 759
     },
     {
       "epoch": 2.9,
       "learning_rate": 1.653944020356234e-06,
-      "loss": 1.5418,
+      "loss": 1.5775,
       "step": 760
     },
     {
       "epoch": 2.9,
       "learning_rate": 1.5903307888040714e-06,
-      "loss": 1.3879,
+      "loss": 1.5154,
       "step": 761
     },
     {
       "epoch": 2.91,
       "learning_rate": 1.5267175572519084e-06,
-      "loss": 1.5063,
+      "loss": 1.4024,
       "step": 762
     },
     {
       "epoch": 2.91,
       "learning_rate": 1.4631043256997457e-06,
-      "loss": 1.4184,
+      "loss": 1.4939,
       "step": 763
     },
     {
       "epoch": 2.92,
       "learning_rate": 1.3994910941475828e-06,
-      "loss": 1.4975,
+      "loss": 1.5221,
       "step": 764
     },
     {
       "epoch": 2.92,
       "learning_rate": 1.3358778625954198e-06,
-      "loss": 1.4759,
+      "loss": 1.5407,
       "step": 765
     },
     {
       "epoch": 2.92,
       "learning_rate": 1.272264631043257e-06,
-      "loss": 1.5125,
+      "loss": 1.5351,
       "step": 766
     },
     {
       "epoch": 2.93,
       "learning_rate": 1.2086513994910942e-06,
-      "loss": 1.462,
+      "loss": 1.4985,
       "step": 767
     },
     {
       "epoch": 2.93,
       "learning_rate": 1.1450381679389313e-06,
-      "loss": 1.4025,
+      "loss": 1.5047,
       "step": 768
     },
     {
       "epoch": 2.94,
       "learning_rate": 1.0814249363867685e-06,
-      "loss": 1.4899,
+      "loss": 1.5272,
       "step": 769
     },
     {
       "epoch": 2.94,
       "learning_rate": 1.0178117048346056e-06,
-      "loss": 1.403,
+      "loss": 1.4864,
       "step": 770
     },
     {
       "epoch": 2.94,
       "learning_rate": 9.54198473282443e-07,
-      "loss": 1.4073,
+      "loss": 1.5018,
       "step": 771
     },
     {
       "epoch": 2.95,
       "learning_rate": 8.9058524173028e-07,
-      "loss": 1.4358,
+      "loss": 1.5671,
       "step": 772
     },
     {
       "epoch": 2.95,
       "learning_rate": 8.26972010178117e-07,
-      "loss": 1.6034,
+      "loss": 1.5599,
       "step": 773
     },
     {
       "epoch": 2.95,
       "learning_rate": 7.633587786259542e-07,
-      "loss": 1.6185,
+      "loss": 1.5463,
       "step": 774
     },
     {
       "epoch": 2.96,
       "learning_rate": 6.997455470737914e-07,
-      "loss": 1.4775,
+      "loss": 1.4985,
       "step": 775
     },
     {
       "epoch": 2.96,
       "learning_rate": 6.361323155216285e-07,
-      "loss": 1.4595,
+      "loss": 1.4801,
       "step": 776
     },
     {
       "epoch": 2.97,
       "learning_rate": 5.725190839694656e-07,
-      "loss": 1.4733,
+      "loss": 1.6203,
       "step": 777
     },
     {
       "epoch": 2.97,
       "learning_rate": 5.089058524173028e-07,
-      "loss": 1.4463,
+      "loss": 1.5507,
       "step": 778
     },
     {
       "epoch": 2.97,
       "learning_rate": 4.4529262086514e-07,
-      "loss": 1.5377,
+      "loss": 1.5613,
       "step": 779
     },
     {
       "epoch": 2.98,
       "learning_rate": 3.816793893129771e-07,
-      "loss": 1.5329,
+      "loss": 1.5681,
       "step": 780
     },
     {
       "epoch": 2.98,
       "learning_rate": 3.1806615776081423e-07,
-      "loss": 1.4827,
+      "loss": 1.4898,
       "step": 781
     },
     {
       "epoch": 2.98,
       "learning_rate": 2.544529262086514e-07,
-      "loss": 1.5151,
+      "loss": 1.5687,
       "step": 782
     },
     {
       "epoch": 2.99,
       "learning_rate": 1.9083969465648855e-07,
-      "loss": 1.5695,
+      "loss": 1.5121,
       "step": 783
     },
     {
       "epoch": 2.99,
       "learning_rate": 1.272264631043257e-07,
-      "loss": 1.4626,
+      "loss": 1.5745,
       "step": 784
     },
     {
       "epoch": 3.0,
       "learning_rate": 6.361323155216285e-08,
-      "loss": 1.5292,
+      "loss": 1.4844,
       "step": 785
     },
     {
       "epoch": 3.0,
       "learning_rate": 0.0,
-      "loss": 1.5321,
+      "loss": 1.3335,
       "step": 786
     },
     {
       "epoch": 3.0,
       "step": 786,
       "total_flos": 1.2699738389348352e+16,
-      "train_loss": 1.6920310260382014,
-      "train_runtime": 827.5296,
-      "train_samples_per_second": 22.734,
-      "train_steps_per_second": 0.95
+      "train_loss": 0.5571378069069549,
+      "train_runtime": 308.6292,
+      "train_samples_per_second": 60.957,
+      "train_steps_per_second": 2.547
     }
   ],
   "max_steps": 786,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99215543b93de901650f1a65728c6e7938885f84434e76915befc1334508cade
+oid sha256:852762d4791077e5ef39a416398ffe5419d39e26caa52ed1b2dde8cb84825133
 size 3567