End of training

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.9900990099009901,
+    "eval_loss": 0.03527973219752312,
+    "eval_runtime": 71.743,
+    "eval_samples_per_second": 29.564,
+    "eval_steps_per_second": 1.854,
+    "total_flos": 1.3545775098737345e+19,
+    "train_loss": 0.23949940906840703,
+    "train_runtime": 11760.3071,
+    "train_samples_per_second": 6.493,
+    "train_steps_per_second": 0.203
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "eval_accuracy": 0.9900990099009901,
+    "eval_loss": 0.03527973219752312,
+    "eval_runtime": 71.743,
+    "eval_samples_per_second": 29.564,
+    "eval_steps_per_second": 1.854
+}

runs/Jul15_23-37-49_0317d6dda330/events.out.tfevents.1689476364.0317d6dda330.905.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:752b25bf2558481fd743b44446b23995094d58d442188455e42b300fcb0f2a95
+size 411

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "total_flos": 1.3545775098737345e+19,
+    "train_loss": 0.23949940906840703,
+    "train_runtime": 11760.3071,
+    "train_samples_per_second": 6.493,
+    "train_steps_per_second": 0.203
+}

trainer_state.json

@@ -0,0 +1,2995 @@
+{
+  "best_metric": 0.9900990099009901,
+  "best_model_checkpoint": "convnextv2-large-1k-224-finetuned-BreastCancer-Classification-BreakHis-AH-60-20-20/checkpoint-2388",
+  "epoch": 12.0,
+  "global_step": 2388,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03,
+      "learning_rate": 1.1627906976744187e-07,
+      "loss": 0.6906,
+      "step": 5
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.3255813953488374e-07,
+      "loss": 0.6925,
+      "step": 10
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 3.4883720930232557e-07,
+      "loss": 0.7,
+      "step": 15
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.651162790697675e-07,
+      "loss": 0.6939,
+      "step": 20
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 5.813953488372093e-07,
+      "loss": 0.6876,
+      "step": 25
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 6.976744186046511e-07,
+      "loss": 0.6907,
+      "step": 30
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 8.13953488372093e-07,
+      "loss": 0.6921,
+      "step": 35
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.30232558139535e-07,
+      "loss": 0.6932,
+      "step": 40
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 1.0465116279069768e-06,
+      "loss": 0.6878,
+      "step": 45
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 1.1627906976744186e-06,
+      "loss": 0.6843,
+      "step": 50
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 1.2790697674418605e-06,
+      "loss": 0.6921,
+      "step": 55
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 1.3953488372093023e-06,
+      "loss": 0.678,
+      "step": 60
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 1.5116279069767443e-06,
+      "loss": 0.6805,
+      "step": 65
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 1.627906976744186e-06,
+      "loss": 0.6728,
+      "step": 70
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 1.744186046511628e-06,
+      "loss": 0.6777,
+      "step": 75
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 1.86046511627907e-06,
+      "loss": 0.6754,
+      "step": 80
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 1.976744186046512e-06,
+      "loss": 0.667,
+      "step": 85
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.0930232558139536e-06,
+      "loss": 0.6691,
+      "step": 90
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 2.2093023255813954e-06,
+      "loss": 0.6699,
+      "step": 95
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 2.325581395348837e-06,
+      "loss": 0.6717,
+      "step": 100
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 2.4418604651162793e-06,
+      "loss": 0.6637,
+      "step": 105
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 2.558139534883721e-06,
+      "loss": 0.6579,
+      "step": 110
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 2.674418604651163e-06,
+      "loss": 0.6467,
+      "step": 115
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 2.7906976744186046e-06,
+      "loss": 0.6505,
+      "step": 120
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 2.9069767441860468e-06,
+      "loss": 0.6403,
+      "step": 125
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.0232558139534885e-06,
+      "loss": 0.6387,
+      "step": 130
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.1395348837209307e-06,
+      "loss": 0.6242,
+      "step": 135
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.255813953488372e-06,
+      "loss": 0.6238,
+      "step": 140
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.3720930232558146e-06,
+      "loss": 0.6152,
+      "step": 145
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 3.488372093023256e-06,
+      "loss": 0.6006,
+      "step": 150
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.604651162790698e-06,
+      "loss": 0.5849,
+      "step": 155
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.72093023255814e-06,
+      "loss": 0.5814,
+      "step": 160
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.837209302325582e-06,
+      "loss": 0.5635,
+      "step": 165
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.953488372093024e-06,
+      "loss": 0.5747,
+      "step": 170
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.0697674418604655e-06,
+      "loss": 0.5792,
+      "step": 175
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.186046511627907e-06,
+      "loss": 0.5577,
+      "step": 180
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.302325581395349e-06,
+      "loss": 0.5225,
+      "step": 185
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.418604651162791e-06,
+      "loss": 0.5204,
+      "step": 190
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.534883720930233e-06,
+      "loss": 0.5207,
+      "step": 195
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8887317303158887,
+      "eval_loss": 0.4744712710380554,
+      "eval_runtime": 820.8676,
+      "eval_samples_per_second": 2.584,
+      "eval_steps_per_second": 0.162,
+      "step": 199
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.651162790697674e-06,
+      "loss": 0.4949,
+      "step": 200
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.767441860465117e-06,
+      "loss": 0.4859,
+      "step": 205
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.883720930232559e-06,
+      "loss": 0.4562,
+      "step": 210
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 5e-06,
+      "loss": 0.4142,
+      "step": 215
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 5.116279069767442e-06,
+      "loss": 0.453,
+      "step": 220
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 5.232558139534884e-06,
+      "loss": 0.424,
+      "step": 225
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 5.348837209302326e-06,
+      "loss": 0.412,
+      "step": 230
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 5.465116279069768e-06,
+      "loss": 0.4098,
+      "step": 235
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 5.581395348837209e-06,
+      "loss": 0.4065,
+      "step": 240
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 5.697674418604652e-06,
+      "loss": 0.4318,
+      "step": 245
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 5.8139534883720935e-06,
+      "loss": 0.3762,
+      "step": 250
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 5.930232558139535e-06,
+      "loss": 0.3467,
+      "step": 255
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 6.046511627906977e-06,
+      "loss": 0.4485,
+      "step": 260
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 6.162790697674419e-06,
+      "loss": 0.3684,
+      "step": 265
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 6.279069767441861e-06,
+      "loss": 0.3518,
+      "step": 270
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 6.395348837209303e-06,
+      "loss": 0.3205,
+      "step": 275
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 6.511627906976744e-06,
+      "loss": 0.2873,
+      "step": 280
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 6.627906976744186e-06,
+      "loss": 0.3188,
+      "step": 285
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 6.744186046511629e-06,
+      "loss": 0.3162,
+      "step": 290
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 6.86046511627907e-06,
+      "loss": 0.2955,
+      "step": 295
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 6.976744186046512e-06,
+      "loss": 0.3233,
+      "step": 300
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 7.093023255813954e-06,
+      "loss": 0.3133,
+      "step": 305
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 7.209302325581396e-06,
+      "loss": 0.2761,
+      "step": 310
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 7.325581395348838e-06,
+      "loss": 0.2793,
+      "step": 315
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 7.44186046511628e-06,
+      "loss": 0.2498,
+      "step": 320
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 7.558139534883721e-06,
+      "loss": 0.2953,
+      "step": 325
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 7.674418604651164e-06,
+      "loss": 0.3078,
+      "step": 330
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 7.790697674418605e-06,
+      "loss": 0.2898,
+      "step": 335
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.906976744186048e-06,
+      "loss": 0.2817,
+      "step": 340
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 8.023255813953488e-06,
+      "loss": 0.246,
+      "step": 345
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 8.139534883720931e-06,
+      "loss": 0.2154,
+      "step": 350
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 8.255813953488372e-06,
+      "loss": 0.2616,
+      "step": 355
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 8.372093023255815e-06,
+      "loss": 0.2622,
+      "step": 360
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 8.488372093023256e-06,
+      "loss": 0.2683,
+      "step": 365
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 8.604651162790698e-06,
+      "loss": 0.2701,
+      "step": 370
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 8.72093023255814e-06,
+      "loss": 0.2216,
+      "step": 375
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 8.837209302325582e-06,
+      "loss": 0.2102,
+      "step": 380
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 8.953488372093023e-06,
+      "loss": 0.1984,
+      "step": 385
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 9.069767441860467e-06,
+      "loss": 0.2258,
+      "step": 390
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 9.186046511627908e-06,
+      "loss": 0.2029,
+      "step": 395
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9401225836869401,
+      "eval_loss": 0.20724786818027496,
+      "eval_runtime": 71.5048,
+      "eval_samples_per_second": 29.662,
+      "eval_steps_per_second": 1.86,
+      "step": 398
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 9.302325581395349e-06,
+      "loss": 0.2323,
+      "step": 400
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 9.418604651162791e-06,
+      "loss": 0.2692,
+      "step": 405
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 9.534883720930234e-06,
+      "loss": 0.2285,
+      "step": 410
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 9.651162790697675e-06,
+      "loss": 0.2077,
+      "step": 415
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 9.767441860465117e-06,
+      "loss": 0.2129,
+      "step": 420
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 9.883720930232558e-06,
+      "loss": 0.1757,
+      "step": 425
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 1e-05,
+      "loss": 0.2437,
+      "step": 430
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 1.0116279069767442e-05,
+      "loss": 0.2745,
+      "step": 435
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 1.0232558139534884e-05,
+      "loss": 0.2724,
+      "step": 440
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.0348837209302325e-05,
+      "loss": 0.1917,
+      "step": 445
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 1.0465116279069768e-05,
+      "loss": 0.1791,
+      "step": 450
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 1.058139534883721e-05,
+      "loss": 0.2349,
+      "step": 455
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 1.0697674418604651e-05,
+      "loss": 0.2102,
+      "step": 460
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 1.0813953488372092e-05,
+      "loss": 0.1248,
+      "step": 465
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 1.0930232558139537e-05,
+      "loss": 0.2014,
+      "step": 470
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 1.1046511627906977e-05,
+      "loss": 0.281,
+      "step": 475
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.1162790697674418e-05,
+      "loss": 0.2636,
+      "step": 480
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.1279069767441861e-05,
+      "loss": 0.1854,
+      "step": 485
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 1.1395348837209304e-05,
+      "loss": 0.2314,
+      "step": 490
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 1.1511627906976744e-05,
+      "loss": 0.2388,
+      "step": 495
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 1.1627906976744187e-05,
+      "loss": 0.2504,
+      "step": 500
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 1.1744186046511628e-05,
+      "loss": 0.1508,
+      "step": 505
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.186046511627907e-05,
+      "loss": 0.1774,
+      "step": 510
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 1.1976744186046513e-05,
+      "loss": 0.1662,
+      "step": 515
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.2093023255813954e-05,
+      "loss": 0.1811,
+      "step": 520
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 1.2209302325581395e-05,
+      "loss": 0.159,
+      "step": 525
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 1.2325581395348838e-05,
+      "loss": 0.1489,
+      "step": 530
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 1.244186046511628e-05,
+      "loss": 0.1532,
+      "step": 535
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 1.2558139534883723e-05,
+      "loss": 0.2108,
+      "step": 540
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 1.2674418604651164e-05,
+      "loss": 0.1863,
+      "step": 545
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 1.2790697674418606e-05,
+      "loss": 0.1417,
+      "step": 550
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 1.2906976744186047e-05,
+      "loss": 0.1903,
+      "step": 555
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 1.3023255813953488e-05,
+      "loss": 0.1849,
+      "step": 560
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.313953488372093e-05,
+      "loss": 0.2094,
+      "step": 565
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.3255813953488372e-05,
+      "loss": 0.2553,
+      "step": 570
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 1.3372093023255814e-05,
+      "loss": 0.1909,
+      "step": 575
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 1.3488372093023258e-05,
+      "loss": 0.1582,
+      "step": 580
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 1.36046511627907e-05,
+      "loss": 0.2467,
+      "step": 585
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 1.372093023255814e-05,
+      "loss": 0.2127,
+      "step": 590
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 1.3837209302325583e-05,
+      "loss": 0.1615,
+      "step": 595
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9547383309759547,
+      "eval_loss": 0.14892685413360596,
+      "eval_runtime": 71.4948,
+      "eval_samples_per_second": 29.666,
+      "eval_steps_per_second": 1.86,
+      "step": 597
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 1.3953488372093024e-05,
+      "loss": 0.1995,
+      "step": 600
+    },
+    {
+      "epoch": 3.04,
+      "learning_rate": 1.4069767441860465e-05,
+      "loss": 0.2305,
+      "step": 605
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 1.4186046511627907e-05,
+      "loss": 0.1808,
+      "step": 610
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.4302325581395348e-05,
+      "loss": 0.1166,
+      "step": 615
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 1.4418604651162792e-05,
+      "loss": 0.1088,
+      "step": 620
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 1.4534883720930233e-05,
+      "loss": 0.2699,
+      "step": 625
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 1.4651162790697676e-05,
+      "loss": 0.1365,
+      "step": 630
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.4767441860465117e-05,
+      "loss": 0.2516,
+      "step": 635
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 1.488372093023256e-05,
+      "loss": 0.2879,
+      "step": 640
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 1.5e-05,
+      "loss": 0.1087,
+      "step": 645
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 1.5116279069767441e-05,
+      "loss": 0.2594,
+      "step": 650
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 1.5232558139534884e-05,
+      "loss": 0.1881,
+      "step": 655
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 1.5348837209302328e-05,
+      "loss": 0.1733,
+      "step": 660
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 1.546511627906977e-05,
+      "loss": 0.1775,
+      "step": 665
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 1.558139534883721e-05,
+      "loss": 0.1316,
+      "step": 670
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 1.569767441860465e-05,
+      "loss": 0.1527,
+      "step": 675
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 1.5813953488372095e-05,
+      "loss": 0.1395,
+      "step": 680
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 1.5930232558139536e-05,
+      "loss": 0.1319,
+      "step": 685
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 1.6046511627906977e-05,
+      "loss": 0.1214,
+      "step": 690
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 1.6162790697674418e-05,
+      "loss": 0.175,
+      "step": 695
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 1.6279069767441862e-05,
+      "loss": 0.1426,
+      "step": 700
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 1.6395348837209303e-05,
+      "loss": 0.238,
+      "step": 705
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 1.6511627906976744e-05,
+      "loss": 0.2168,
+      "step": 710
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 1.6627906976744188e-05,
+      "loss": 0.2221,
+      "step": 715
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 1.674418604651163e-05,
+      "loss": 0.1813,
+      "step": 720
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 1.686046511627907e-05,
+      "loss": 0.0943,
+      "step": 725
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 1.697674418604651e-05,
+      "loss": 0.1289,
+      "step": 730
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 1.7093023255813952e-05,
+      "loss": 0.2804,
+      "step": 735
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 1.7209302325581396e-05,
+      "loss": 0.2284,
+      "step": 740
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 1.732558139534884e-05,
+      "loss": 0.1363,
+      "step": 745
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 1.744186046511628e-05,
+      "loss": 0.6221,
+      "step": 750
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 1.7558139534883722e-05,
+      "loss": 0.408,
+      "step": 755
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 1.7674418604651163e-05,
+      "loss": 0.1856,
+      "step": 760
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 1.7790697674418604e-05,
+      "loss": 0.1677,
+      "step": 765
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 1.7906976744186045e-05,
+      "loss": 0.1393,
+      "step": 770
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 1.802325581395349e-05,
+      "loss": 0.2656,
+      "step": 775
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 1.8139534883720934e-05,
+      "loss": 0.0915,
+      "step": 780
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 1.8255813953488375e-05,
+      "loss": 0.1772,
+      "step": 785
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 1.8372093023255815e-05,
+      "loss": 0.3261,
+      "step": 790
+    },
+    {
+      "epoch": 3.99,
+      "learning_rate": 1.8488372093023256e-05,
+      "loss": 0.1662,
+      "step": 795
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9561527581329562,
+      "eval_loss": 0.13120350241661072,
+      "eval_runtime": 71.6187,
+      "eval_samples_per_second": 29.615,
+      "eval_steps_per_second": 1.857,
+      "step": 796
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 1.8604651162790697e-05,
+      "loss": 0.1648,
+      "step": 800
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 1.8720930232558138e-05,
+      "loss": 0.2065,
+      "step": 805
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 1.8837209302325582e-05,
+      "loss": 0.1699,
+      "step": 810
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 1.8953488372093023e-05,
+      "loss": 0.1036,
+      "step": 815
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 1.9069767441860468e-05,
+      "loss": 0.0774,
+      "step": 820
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 1.918604651162791e-05,
+      "loss": 0.2527,
+      "step": 825
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 1.930232558139535e-05,
+      "loss": 0.1152,
+      "step": 830
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 1.941860465116279e-05,
+      "loss": 0.1183,
+      "step": 835
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 1.9534883720930235e-05,
+      "loss": 0.2491,
+      "step": 840
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 1.9651162790697676e-05,
+      "loss": 0.1142,
+      "step": 845
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 1.9767441860465116e-05,
+      "loss": 0.0765,
+      "step": 850
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 1.9883720930232557e-05,
+      "loss": 0.1876,
+      "step": 855
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 2e-05,
+      "loss": 0.1759,
+      "step": 860
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 2.0116279069767443e-05,
+      "loss": 0.0911,
+      "step": 865
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 2.0232558139534883e-05,
+      "loss": 0.2646,
+      "step": 870
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 2.0348837209302328e-05,
+      "loss": 0.0842,
+      "step": 875
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 2.046511627906977e-05,
+      "loss": 0.2088,
+      "step": 880
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 2.058139534883721e-05,
+      "loss": 0.1186,
+      "step": 885
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 2.069767441860465e-05,
+      "loss": 0.0894,
+      "step": 890
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 2.081395348837209e-05,
+      "loss": 0.1024,
+      "step": 895
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 2.0930232558139536e-05,
+      "loss": 0.1094,
+      "step": 900
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 2.104651162790698e-05,
+      "loss": 0.1731,
+      "step": 905
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 2.116279069767442e-05,
+      "loss": 0.2125,
+      "step": 910
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 2.1279069767441862e-05,
+      "loss": 0.1469,
+      "step": 915
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 2.1395348837209303e-05,
+      "loss": 0.1238,
+      "step": 920
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 2.1511627906976744e-05,
+      "loss": 0.1013,
+      "step": 925
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 2.1627906976744184e-05,
+      "loss": 0.1775,
+      "step": 930
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 2.174418604651163e-05,
+      "loss": 0.1207,
+      "step": 935
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 2.1860465116279073e-05,
+      "loss": 0.0734,
+      "step": 940
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 2.1976744186046514e-05,
+      "loss": 0.1425,
+      "step": 945
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 2.2093023255813955e-05,
+      "loss": 0.0941,
+      "step": 950
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 2.2209302325581396e-05,
+      "loss": 0.1594,
+      "step": 955
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 2.2325581395348837e-05,
+      "loss": 0.142,
+      "step": 960
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 2.244186046511628e-05,
+      "loss": 0.1971,
+      "step": 965
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 2.2558139534883722e-05,
+      "loss": 0.1717,
+      "step": 970
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 2.2674418604651163e-05,
+      "loss": 0.1267,
+      "step": 975
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 2.2790697674418607e-05,
+      "loss": 0.1118,
+      "step": 980
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 2.2906976744186048e-05,
+      "loss": 0.132,
+      "step": 985
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 2.302325581395349e-05,
+      "loss": 0.225,
+      "step": 990
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.313953488372093e-05,
+      "loss": 0.1986,
+      "step": 995
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9698255539839699,
+      "eval_loss": 0.10262064635753632,
+      "eval_runtime": 71.6971,
+      "eval_samples_per_second": 29.583,
+      "eval_steps_per_second": 1.855,
+      "step": 995
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 2.3255813953488374e-05,
+      "loss": 0.1295,
+      "step": 1000
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 2.3372093023255815e-05,
+      "loss": 0.1204,
+      "step": 1005
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 2.3488372093023256e-05,
+      "loss": 0.1205,
+      "step": 1010
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 2.3604651162790697e-05,
+      "loss": 0.1513,
+      "step": 1015
+    },
+    {
+      "epoch": 5.13,
+      "learning_rate": 2.372093023255814e-05,
+      "loss": 0.2574,
+      "step": 1020
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 2.3837209302325582e-05,
+      "loss": 0.2257,
+      "step": 1025
+    },
+    {
+      "epoch": 5.18,
+      "learning_rate": 2.3953488372093026e-05,
+      "loss": 0.0617,
+      "step": 1030
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 2.4069767441860467e-05,
+      "loss": 0.1194,
+      "step": 1035
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 2.4186046511627908e-05,
+      "loss": 0.1783,
+      "step": 1040
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 2.430232558139535e-05,
+      "loss": 0.0697,
+      "step": 1045
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 2.441860465116279e-05,
+      "loss": 0.1334,
+      "step": 1050
+    },
+    {
+      "epoch": 5.3,
+      "learning_rate": 2.453488372093023e-05,
+      "loss": 0.0817,
+      "step": 1055
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 2.4651162790697675e-05,
+      "loss": 0.1887,
+      "step": 1060
+    },
+    {
+      "epoch": 5.35,
+      "learning_rate": 2.476744186046512e-05,
+      "loss": 0.1225,
+      "step": 1065
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 2.488372093023256e-05,
+      "loss": 0.168,
+      "step": 1070
+    },
+    {
+      "epoch": 5.4,
+      "learning_rate": 2.5e-05,
+      "loss": 0.1217,
+      "step": 1075
+    },
+    {
+      "epoch": 5.43,
+      "learning_rate": 2.5116279069767445e-05,
+      "loss": 0.1142,
+      "step": 1080
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 2.5232558139534883e-05,
+      "loss": 0.1478,
+      "step": 1085
+    },
+    {
+      "epoch": 5.48,
+      "learning_rate": 2.5348837209302327e-05,
+      "loss": 0.1616,
+      "step": 1090
+    },
+    {
+      "epoch": 5.5,
+      "learning_rate": 2.5465116279069768e-05,
+      "loss": 0.1622,
+      "step": 1095
+    },
+    {
+      "epoch": 5.53,
+      "learning_rate": 2.5581395348837212e-05,
+      "loss": 0.1199,
+      "step": 1100
+    },
+    {
+      "epoch": 5.55,
+      "learning_rate": 2.569767441860465e-05,
+      "loss": 0.1247,
+      "step": 1105
+    },
+    {
+      "epoch": 5.58,
+      "learning_rate": 2.5813953488372094e-05,
+      "loss": 0.0762,
+      "step": 1110
+    },
+    {
+      "epoch": 5.6,
+      "learning_rate": 2.593023255813954e-05,
+      "loss": 0.0909,
+      "step": 1115
+    },
+    {
+      "epoch": 5.63,
+      "learning_rate": 2.6046511627906976e-05,
+      "loss": 0.1148,
+      "step": 1120
+    },
+    {
+      "epoch": 5.65,
+      "learning_rate": 2.616279069767442e-05,
+      "loss": 0.1652,
+      "step": 1125
+    },
+    {
+      "epoch": 5.68,
+      "learning_rate": 2.627906976744186e-05,
+      "loss": 0.0933,
+      "step": 1130
+    },
+    {
+      "epoch": 5.7,
+      "learning_rate": 2.6395348837209306e-05,
+      "loss": 0.1661,
+      "step": 1135
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 2.6511627906976743e-05,
+      "loss": 0.1631,
+      "step": 1140
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 2.6627906976744187e-05,
+      "loss": 0.0477,
+      "step": 1145
+    },
+    {
+      "epoch": 5.78,
+      "learning_rate": 2.674418604651163e-05,
+      "loss": 0.1101,
+      "step": 1150
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 2.6860465116279073e-05,
+      "loss": 0.1127,
+      "step": 1155
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 2.6976744186046517e-05,
+      "loss": 0.0872,
+      "step": 1160
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 2.7093023255813954e-05,
+      "loss": 0.1027,
+      "step": 1165
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 2.72093023255814e-05,
+      "loss": 0.0929,
+      "step": 1170
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 2.7325581395348836e-05,
+      "loss": 0.0588,
+      "step": 1175
+    },
+    {
+      "epoch": 5.93,
+      "learning_rate": 2.744186046511628e-05,
+      "loss": 0.0911,
+      "step": 1180
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 2.755813953488372e-05,
+      "loss": 0.0948,
+      "step": 1185
+    },
+    {
+      "epoch": 5.98,
+      "learning_rate": 2.7674418604651166e-05,
+      "loss": 0.0854,
+      "step": 1190
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9801980198019802,
+      "eval_loss": 0.05832900106906891,
+      "eval_runtime": 71.4977,
+      "eval_samples_per_second": 29.665,
+      "eval_steps_per_second": 1.86,
+      "step": 1194
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 2.779069767441861e-05,
+      "loss": 0.1528,
+      "step": 1195
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 2.7906976744186048e-05,
+      "loss": 0.0916,
+      "step": 1200
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 2.8023255813953492e-05,
+      "loss": 0.1095,
+      "step": 1205
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 2.813953488372093e-05,
+      "loss": 0.0748,
+      "step": 1210
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 2.8255813953488374e-05,
+      "loss": 0.1212,
+      "step": 1215
+    },
+    {
+      "epoch": 6.13,
+      "learning_rate": 2.8372093023255815e-05,
+      "loss": 0.1115,
+      "step": 1220
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 2.848837209302326e-05,
+      "loss": 0.0926,
+      "step": 1225
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 2.8604651162790696e-05,
+      "loss": 0.1111,
+      "step": 1230
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 2.872093023255814e-05,
+      "loss": 0.0805,
+      "step": 1235
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 2.8837209302325585e-05,
+      "loss": 0.0785,
+      "step": 1240
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 2.8953488372093022e-05,
+      "loss": 0.1897,
+      "step": 1245
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 2.9069767441860467e-05,
+      "loss": 0.1967,
+      "step": 1250
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 2.9186046511627908e-05,
+      "loss": 0.1282,
+      "step": 1255
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 2.9302325581395352e-05,
+      "loss": 0.055,
+      "step": 1260
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 2.941860465116279e-05,
+      "loss": 0.071,
+      "step": 1265
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 2.9534883720930234e-05,
+      "loss": 0.1448,
+      "step": 1270
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 2.9651162790697678e-05,
+      "loss": 0.189,
+      "step": 1275
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 2.976744186046512e-05,
+      "loss": 0.0417,
+      "step": 1280
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 2.9883720930232563e-05,
+      "loss": 0.1256,
+      "step": 1285
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 3e-05,
+      "loss": 0.0677,
+      "step": 1290
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 3.0116279069767445e-05,
+      "loss": 0.2087,
+      "step": 1295
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 3.0232558139534883e-05,
+      "loss": 0.1094,
+      "step": 1300
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 3.0348837209302327e-05,
+      "loss": 0.097,
+      "step": 1305
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 3.0465116279069768e-05,
+      "loss": 0.1087,
+      "step": 1310
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 3.058139534883721e-05,
+      "loss": 0.0655,
+      "step": 1315
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 3.0697674418604656e-05,
+      "loss": 0.2398,
+      "step": 1320
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 3.081395348837209e-05,
+      "loss": 0.0626,
+      "step": 1325
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 3.093023255813954e-05,
+      "loss": 0.1554,
+      "step": 1330
+    },
+    {
+      "epoch": 6.71,
+      "learning_rate": 3.104651162790698e-05,
+      "loss": 0.0479,
+      "step": 1335
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 3.116279069767442e-05,
+      "loss": 0.1825,
+      "step": 1340
+    },
+    {
+      "epoch": 6.76,
+      "learning_rate": 3.127906976744186e-05,
+      "loss": 0.2722,
+      "step": 1345
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 3.13953488372093e-05,
+      "loss": 0.0999,
+      "step": 1350
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 3.151162790697675e-05,
+      "loss": 0.1423,
+      "step": 1355
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 3.162790697674419e-05,
+      "loss": 0.1435,
+      "step": 1360
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 3.174418604651163e-05,
+      "loss": 0.1321,
+      "step": 1365
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 3.186046511627907e-05,
+      "loss": 0.0719,
+      "step": 1370
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 3.197674418604651e-05,
+      "loss": 0.1688,
+      "step": 1375
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 3.2093023255813954e-05,
+      "loss": 0.0583,
+      "step": 1380
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 3.2209302325581395e-05,
+      "loss": 0.1247,
+      "step": 1385
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 3.2325581395348836e-05,
+      "loss": 0.0538,
+      "step": 1390
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9834983498349835,
+      "eval_loss": 0.056816570460796356,
+      "eval_runtime": 71.6829,
+      "eval_samples_per_second": 29.589,
+      "eval_steps_per_second": 1.855,
+      "step": 1393
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 3.2441860465116283e-05,
+      "loss": 0.1301,
+      "step": 1395
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 3.2558139534883724e-05,
+      "loss": 0.078,
+      "step": 1400
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 3.2674418604651165e-05,
+      "loss": 0.121,
+      "step": 1405
+    },
+    {
+      "epoch": 7.09,
+      "learning_rate": 3.2790697674418606e-05,
+      "loss": 0.4797,
+      "step": 1410
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 3.290697674418605e-05,
+      "loss": 0.3881,
+      "step": 1415
+    },
+    {
+      "epoch": 7.14,
+      "learning_rate": 3.302325581395349e-05,
+      "loss": 0.3787,
+      "step": 1420
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 3.313953488372093e-05,
+      "loss": 0.3241,
+      "step": 1425
+    },
+    {
+      "epoch": 7.19,
+      "learning_rate": 3.3255813953488377e-05,
+      "loss": 0.1639,
+      "step": 1430
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 3.337209302325582e-05,
+      "loss": 0.2431,
+      "step": 1435
+    },
+    {
+      "epoch": 7.24,
+      "learning_rate": 3.348837209302326e-05,
+      "loss": 0.1114,
+      "step": 1440
+    },
+    {
+      "epoch": 7.26,
+      "learning_rate": 3.36046511627907e-05,
+      "loss": 0.0814,
+      "step": 1445
+    },
+    {
+      "epoch": 7.29,
+      "learning_rate": 3.372093023255814e-05,
+      "loss": 0.1274,
+      "step": 1450
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 3.383720930232558e-05,
+      "loss": 0.1317,
+      "step": 1455
+    },
+    {
+      "epoch": 7.34,
+      "learning_rate": 3.395348837209302e-05,
+      "loss": 0.2124,
+      "step": 1460
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 3.406976744186047e-05,
+      "loss": 0.1573,
+      "step": 1465
+    },
+    {
+      "epoch": 7.39,
+      "learning_rate": 3.4186046511627904e-05,
+      "loss": 0.0797,
+      "step": 1470
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 3.430232558139535e-05,
+      "loss": 0.0944,
+      "step": 1475
+    },
+    {
+      "epoch": 7.44,
+      "learning_rate": 3.441860465116279e-05,
+      "loss": 0.1133,
+      "step": 1480
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 3.453488372093023e-05,
+      "loss": 0.0567,
+      "step": 1485
+    },
+    {
+      "epoch": 7.49,
+      "learning_rate": 3.465116279069768e-05,
+      "loss": 0.0431,
+      "step": 1490
+    },
+    {
+      "epoch": 7.51,
+      "learning_rate": 3.4767441860465115e-05,
+      "loss": 0.0645,
+      "step": 1495
+    },
+    {
+      "epoch": 7.54,
+      "learning_rate": 3.488372093023256e-05,
+      "loss": 0.1273,
+      "step": 1500
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 3.5e-05,
+      "loss": 0.0888,
+      "step": 1505
+    },
+    {
+      "epoch": 7.59,
+      "learning_rate": 3.5116279069767445e-05,
+      "loss": 0.0734,
+      "step": 1510
+    },
+    {
+      "epoch": 7.61,
+      "learning_rate": 3.5232558139534886e-05,
+      "loss": 0.119,
+      "step": 1515
+    },
+    {
+      "epoch": 7.64,
+      "learning_rate": 3.5348837209302326e-05,
+      "loss": 0.094,
+      "step": 1520
+    },
+    {
+      "epoch": 7.66,
+      "learning_rate": 3.5465116279069774e-05,
+      "loss": 0.1113,
+      "step": 1525
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 3.558139534883721e-05,
+      "loss": 0.1639,
+      "step": 1530
+    },
+    {
+      "epoch": 7.71,
+      "learning_rate": 3.5697674418604656e-05,
+      "loss": 0.1105,
+      "step": 1535
+    },
+    {
+      "epoch": 7.74,
+      "learning_rate": 3.581395348837209e-05,
+      "loss": 0.1542,
+      "step": 1540
+    },
+    {
+      "epoch": 7.76,
+      "learning_rate": 3.593023255813954e-05,
+      "loss": 0.1406,
+      "step": 1545
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 3.604651162790698e-05,
+      "loss": 0.0928,
+      "step": 1550
+    },
+    {
+      "epoch": 7.81,
+      "learning_rate": 3.616279069767442e-05,
+      "loss": 0.1334,
+      "step": 1555
+    },
+    {
+      "epoch": 7.84,
+      "learning_rate": 3.627906976744187e-05,
+      "loss": 0.1222,
+      "step": 1560
+    },
+    {
+      "epoch": 7.86,
+      "learning_rate": 3.63953488372093e-05,
+      "loss": 0.0813,
+      "step": 1565
+    },
+    {
+      "epoch": 7.89,
+      "learning_rate": 3.651162790697675e-05,
+      "loss": 0.0702,
+      "step": 1570
+    },
+    {
+      "epoch": 7.91,
+      "learning_rate": 3.662790697674418e-05,
+      "loss": 0.06,
+      "step": 1575
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 3.674418604651163e-05,
+      "loss": 0.1439,
+      "step": 1580
+    },
+    {
+      "epoch": 7.96,
+      "learning_rate": 3.686046511627907e-05,
+      "loss": 0.1491,
+      "step": 1585
+    },
+    {
+      "epoch": 7.99,
+      "learning_rate": 3.697674418604651e-05,
+      "loss": 0.0977,
+      "step": 1590
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9792550683639792,
+      "eval_loss": 0.06544233113527298,
+      "eval_runtime": 71.7179,
+      "eval_samples_per_second": 29.574,
+      "eval_steps_per_second": 1.854,
+      "step": 1592
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 3.709302325581396e-05,
+      "loss": 0.0432,
+      "step": 1595
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 3.7209302325581394e-05,
+      "loss": 0.21,
+      "step": 1600
+    },
+    {
+      "epoch": 8.07,
+      "learning_rate": 3.732558139534884e-05,
+      "loss": 0.1797,
+      "step": 1605
+    },
+    {
+      "epoch": 8.09,
+      "learning_rate": 3.7441860465116276e-05,
+      "loss": 0.0552,
+      "step": 1610
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 3.7558139534883724e-05,
+      "loss": 0.217,
+      "step": 1615
+    },
+    {
+      "epoch": 8.14,
+      "learning_rate": 3.7674418604651165e-05,
+      "loss": 0.2101,
+      "step": 1620
+    },
+    {
+      "epoch": 8.17,
+      "learning_rate": 3.7790697674418606e-05,
+      "loss": 0.2016,
+      "step": 1625
+    },
+    {
+      "epoch": 8.19,
+      "learning_rate": 3.790697674418605e-05,
+      "loss": 0.1252,
+      "step": 1630
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 3.802325581395349e-05,
+      "loss": 0.1332,
+      "step": 1635
+    },
+    {
+      "epoch": 8.24,
+      "learning_rate": 3.8139534883720935e-05,
+      "loss": 0.1404,
+      "step": 1640
+    },
+    {
+      "epoch": 8.27,
+      "learning_rate": 3.8255813953488376e-05,
+      "loss": 0.2552,
+      "step": 1645
+    },
+    {
+      "epoch": 8.29,
+      "learning_rate": 3.837209302325582e-05,
+      "loss": 0.2006,
+      "step": 1650
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 3.848837209302326e-05,
+      "loss": 0.1864,
+      "step": 1655
+    },
+    {
+      "epoch": 8.34,
+      "learning_rate": 3.86046511627907e-05,
+      "loss": 0.2003,
+      "step": 1660
+    },
+    {
+      "epoch": 8.37,
+      "learning_rate": 3.872093023255814e-05,
+      "loss": 0.2577,
+      "step": 1665
+    },
+    {
+      "epoch": 8.39,
+      "learning_rate": 3.883720930232558e-05,
+      "loss": 0.1968,
+      "step": 1670
+    },
+    {
+      "epoch": 8.42,
+      "learning_rate": 3.895348837209303e-05,
+      "loss": 0.1838,
+      "step": 1675
+    },
+    {
+      "epoch": 8.44,
+      "learning_rate": 3.906976744186047e-05,
+      "loss": 0.1788,
+      "step": 1680
+    },
+    {
+      "epoch": 8.47,
+      "learning_rate": 3.918604651162791e-05,
+      "loss": 0.0573,
+      "step": 1685
+    },
+    {
+      "epoch": 8.49,
+      "learning_rate": 3.930232558139535e-05,
+      "loss": 0.0858,
+      "step": 1690
+    },
+    {
+      "epoch": 8.52,
+      "learning_rate": 3.941860465116279e-05,
+      "loss": 0.0236,
+      "step": 1695
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 3.953488372093023e-05,
+      "loss": 0.043,
+      "step": 1700
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 3.9651162790697674e-05,
+      "loss": 0.1477,
+      "step": 1705
+    },
+    {
+      "epoch": 8.59,
+      "learning_rate": 3.9767441860465115e-05,
+      "loss": 0.2065,
+      "step": 1710
+    },
+    {
+      "epoch": 8.62,
+      "learning_rate": 3.988372093023256e-05,
+      "loss": 0.3585,
+      "step": 1715
+    },
+    {
+      "epoch": 8.64,
+      "learning_rate": 4e-05,
+      "loss": 0.538,
+      "step": 1720
+    },
+    {
+      "epoch": 8.67,
+      "learning_rate": 4.0116279069767444e-05,
+      "loss": 0.7162,
+      "step": 1725
+    },
+    {
+      "epoch": 8.69,
+      "learning_rate": 4.0232558139534885e-05,
+      "loss": 0.6912,
+      "step": 1730
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 4.0348837209302326e-05,
+      "loss": 0.7057,
+      "step": 1735
+    },
+    {
+      "epoch": 8.74,
+      "learning_rate": 4.046511627906977e-05,
+      "loss": 0.6983,
+      "step": 1740
+    },
+    {
+      "epoch": 8.77,
+      "learning_rate": 4.058139534883721e-05,
+      "loss": 0.6928,
+      "step": 1745
+    },
+    {
+      "epoch": 8.79,
+      "learning_rate": 4.0697674418604655e-05,
+      "loss": 0.6986,
+      "step": 1750
+    },
+    {
+      "epoch": 8.82,
+      "learning_rate": 4.0813953488372096e-05,
+      "loss": 0.6948,
+      "step": 1755
+    },
+    {
+      "epoch": 8.84,
+      "learning_rate": 4.093023255813954e-05,
+      "loss": 0.694,
+      "step": 1760
+    },
+    {
+      "epoch": 8.87,
+      "learning_rate": 4.104651162790698e-05,
+      "loss": 0.7074,
+      "step": 1765
+    },
+    {
+      "epoch": 8.89,
+      "learning_rate": 4.116279069767442e-05,
+      "loss": 0.6922,
+      "step": 1770
+    },
+    {
+      "epoch": 8.92,
+      "learning_rate": 4.127906976744187e-05,
+      "loss": 0.6914,
+      "step": 1775
+    },
+    {
+      "epoch": 8.94,
+      "learning_rate": 4.13953488372093e-05,
+      "loss": 0.6987,
+      "step": 1780
+    },
+    {
+      "epoch": 8.97,
+      "learning_rate": 4.151162790697675e-05,
+      "loss": 0.6934,
+      "step": 1785
+    },
+    {
+      "epoch": 8.99,
+      "learning_rate": 4.162790697674418e-05,
+      "loss": 0.6971,
+      "step": 1790
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.545025931164545,
+      "eval_loss": 0.6821029782295227,
+      "eval_runtime": 71.3695,
+      "eval_samples_per_second": 29.719,
+      "eval_steps_per_second": 1.864,
+      "step": 1791
+    },
+    {
+      "epoch": 9.02,
+      "learning_rate": 4.174418604651163e-05,
+      "loss": 0.6789,
+      "step": 1795
+    },
+    {
+      "epoch": 9.05,
+      "learning_rate": 4.186046511627907e-05,
+      "loss": 0.6792,
+      "step": 1800
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 4.197674418604651e-05,
+      "loss": 0.6635,
+      "step": 1805
+    },
+    {
+      "epoch": 9.1,
+      "learning_rate": 4.209302325581396e-05,
+      "loss": 0.6199,
+      "step": 1810
+    },
+    {
+      "epoch": 9.12,
+      "learning_rate": 4.2209302325581394e-05,
+      "loss": 0.5559,
+      "step": 1815
+    },
+    {
+      "epoch": 9.15,
+      "learning_rate": 4.232558139534884e-05,
+      "loss": 0.6427,
+      "step": 1820
+    },
+    {
+      "epoch": 9.17,
+      "learning_rate": 4.2441860465116276e-05,
+      "loss": 0.5806,
+      "step": 1825
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 4.2558139534883724e-05,
+      "loss": 0.5822,
+      "step": 1830
+    },
+    {
+      "epoch": 9.22,
+      "learning_rate": 4.2674418604651164e-05,
+      "loss": 0.6987,
+      "step": 1835
+    },
+    {
+      "epoch": 9.25,
+      "learning_rate": 4.2790697674418605e-05,
+      "loss": 0.5803,
+      "step": 1840
+    },
+    {
+      "epoch": 9.27,
+      "learning_rate": 4.290697674418605e-05,
+      "loss": 0.6675,
+      "step": 1845
+    },
+    {
+      "epoch": 9.3,
+      "learning_rate": 4.302325581395349e-05,
+      "loss": 0.4708,
+      "step": 1850
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 4.3139534883720935e-05,
+      "loss": 0.2945,
+      "step": 1855
+    },
+    {
+      "epoch": 9.35,
+      "learning_rate": 4.325581395348837e-05,
+      "loss": 0.3822,
+      "step": 1860
+    },
+    {
+      "epoch": 9.37,
+      "learning_rate": 4.337209302325582e-05,
+      "loss": 0.372,
+      "step": 1865
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 4.348837209302326e-05,
+      "loss": 0.346,
+      "step": 1870
+    },
+    {
+      "epoch": 9.42,
+      "learning_rate": 4.36046511627907e-05,
+      "loss": 0.3051,
+      "step": 1875
+    },
+    {
+      "epoch": 9.45,
+      "learning_rate": 4.3720930232558146e-05,
+      "loss": 0.4088,
+      "step": 1880
+    },
+    {
+      "epoch": 9.47,
+      "learning_rate": 4.383720930232558e-05,
+      "loss": 0.477,
+      "step": 1885
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 4.395348837209303e-05,
+      "loss": 0.4706,
+      "step": 1890
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 4.406976744186047e-05,
+      "loss": 0.5401,
+      "step": 1895
+    },
+    {
+      "epoch": 9.55,
+      "learning_rate": 4.418604651162791e-05,
+      "loss": 0.414,
+      "step": 1900
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 4.430232558139535e-05,
+      "loss": 0.3895,
+      "step": 1905
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 4.441860465116279e-05,
+      "loss": 0.294,
+      "step": 1910
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 4.453488372093024e-05,
+      "loss": 0.1859,
+      "step": 1915
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 4.465116279069767e-05,
+      "loss": 0.208,
+      "step": 1920
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 4.476744186046512e-05,
+      "loss": 0.1935,
+      "step": 1925
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 4.488372093023256e-05,
+      "loss": 0.1841,
+      "step": 1930
+    },
+    {
+      "epoch": 9.72,
+      "learning_rate": 4.5e-05,
+      "loss": 0.1984,
+      "step": 1935
+    },
+    {
+      "epoch": 9.75,
+      "learning_rate": 4.5116279069767444e-05,
+      "loss": 0.1826,
+      "step": 1940
+    },
+    {
+      "epoch": 9.77,
+      "learning_rate": 4.5232558139534885e-05,
+      "loss": 0.1915,
+      "step": 1945
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 4.5348837209302326e-05,
+      "loss": 0.1121,
+      "step": 1950
+    },
+    {
+      "epoch": 9.82,
+      "learning_rate": 4.5465116279069766e-05,
+      "loss": 0.1475,
+      "step": 1955
+    },
+    {
+      "epoch": 9.85,
+      "learning_rate": 4.5581395348837214e-05,
+      "loss": 0.0673,
+      "step": 1960
+    },
+    {
+      "epoch": 9.87,
+      "learning_rate": 4.5697674418604655e-05,
+      "loss": 0.1644,
+      "step": 1965
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 4.5813953488372096e-05,
+      "loss": 0.2278,
+      "step": 1970
+    },
+    {
+      "epoch": 9.92,
+      "learning_rate": 4.593023255813954e-05,
+      "loss": 0.1911,
+      "step": 1975
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 4.604651162790698e-05,
+      "loss": 0.2795,
+      "step": 1980
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 4.616279069767442e-05,
+      "loss": 0.125,
+      "step": 1985
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 4.627906976744186e-05,
+      "loss": 0.211,
+      "step": 1990
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9325789721829326,
+      "eval_loss": 0.16544359922409058,
+      "eval_runtime": 71.699,
+      "eval_samples_per_second": 29.582,
+      "eval_steps_per_second": 1.855,
+      "step": 1990
+    },
+    {
+      "epoch": 10.03,
+      "learning_rate": 4.639534883720931e-05,
+      "loss": 0.1237,
+      "step": 1995
+    },
+    {
+      "epoch": 10.05,
+      "learning_rate": 4.651162790697675e-05,
+      "loss": 0.2095,
+      "step": 2000
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 4.662790697674419e-05,
+      "loss": 0.0815,
+      "step": 2005
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 4.674418604651163e-05,
+      "loss": 0.0781,
+      "step": 2010
+    },
+    {
+      "epoch": 10.13,
+      "learning_rate": 4.686046511627907e-05,
+      "loss": 0.0592,
+      "step": 2015
+    },
+    {
+      "epoch": 10.15,
+      "learning_rate": 4.697674418604651e-05,
+      "loss": 0.1002,
+      "step": 2020
+    },
+    {
+      "epoch": 10.18,
+      "learning_rate": 4.709302325581396e-05,
+      "loss": 0.2596,
+      "step": 2025
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 4.7209302325581394e-05,
+      "loss": 0.2687,
+      "step": 2030
+    },
+    {
+      "epoch": 10.23,
+      "learning_rate": 4.732558139534884e-05,
+      "loss": 0.3895,
+      "step": 2035
+    },
+    {
+      "epoch": 10.25,
+      "learning_rate": 4.744186046511628e-05,
+      "loss": 0.176,
+      "step": 2040
+    },
+    {
+      "epoch": 10.28,
+      "learning_rate": 4.755813953488372e-05,
+      "loss": 0.2329,
+      "step": 2045
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 4.7674418604651164e-05,
+      "loss": 0.1346,
+      "step": 2050
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 4.7790697674418605e-05,
+      "loss": 0.1283,
+      "step": 2055
+    },
+    {
+      "epoch": 10.35,
+      "learning_rate": 4.790697674418605e-05,
+      "loss": 0.1299,
+      "step": 2060
+    },
+    {
+      "epoch": 10.38,
+      "learning_rate": 4.802325581395349e-05,
+      "loss": 0.1114,
+      "step": 2065
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 4.8139534883720934e-05,
+      "loss": 0.1341,
+      "step": 2070
+    },
+    {
+      "epoch": 10.43,
+      "learning_rate": 4.8255813953488375e-05,
+      "loss": 0.1274,
+      "step": 2075
+    },
+    {
+      "epoch": 10.45,
+      "learning_rate": 4.8372093023255816e-05,
+      "loss": 0.0595,
+      "step": 2080
+    },
+    {
+      "epoch": 10.48,
+      "learning_rate": 4.848837209302326e-05,
+      "loss": 0.0398,
+      "step": 2085
+    },
+    {
+      "epoch": 10.5,
+      "learning_rate": 4.86046511627907e-05,
+      "loss": 0.1733,
+      "step": 2090
+    },
+    {
+      "epoch": 10.53,
+      "learning_rate": 4.8720930232558146e-05,
+      "loss": 0.1836,
+      "step": 2095
+    },
+    {
+      "epoch": 10.55,
+      "learning_rate": 4.883720930232558e-05,
+      "loss": 0.0654,
+      "step": 2100
+    },
+    {
+      "epoch": 10.58,
+      "learning_rate": 4.895348837209303e-05,
+      "loss": 0.0939,
+      "step": 2105
+    },
+    {
+      "epoch": 10.6,
+      "learning_rate": 4.906976744186046e-05,
+      "loss": 0.1409,
+      "step": 2110
+    },
+    {
+      "epoch": 10.63,
+      "learning_rate": 4.918604651162791e-05,
+      "loss": 0.1129,
+      "step": 2115
+    },
+    {
+      "epoch": 10.65,
+      "learning_rate": 4.930232558139535e-05,
+      "loss": 0.3178,
+      "step": 2120
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 4.941860465116279e-05,
+      "loss": 0.1593,
+      "step": 2125
+    },
+    {
+      "epoch": 10.7,
+      "learning_rate": 4.953488372093024e-05,
+      "loss": 0.0817,
+      "step": 2130
+    },
+    {
+      "epoch": 10.73,
+      "learning_rate": 4.965116279069767e-05,
+      "loss": 0.1922,
+      "step": 2135
+    },
+    {
+      "epoch": 10.75,
+      "learning_rate": 4.976744186046512e-05,
+      "loss": 0.0615,
+      "step": 2140
+    },
+    {
+      "epoch": 10.78,
+      "learning_rate": 4.9883720930232555e-05,
+      "loss": 0.4146,
+      "step": 2145
+    },
+    {
+      "epoch": 10.8,
+      "learning_rate": 5e-05,
+      "loss": 0.3568,
+      "step": 2150
+    },
+    {
+      "epoch": 10.83,
+      "learning_rate": 4.8949579831932775e-05,
+      "loss": 0.0982,
+      "step": 2155
+    },
+    {
+      "epoch": 10.85,
+      "learning_rate": 4.7899159663865554e-05,
+      "loss": 0.1187,
+      "step": 2160
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 4.684873949579832e-05,
+      "loss": 0.0911,
+      "step": 2165
+    },
+    {
+      "epoch": 10.9,
+      "learning_rate": 4.579831932773109e-05,
+      "loss": 0.1554,
+      "step": 2170
+    },
+    {
+      "epoch": 10.93,
+      "learning_rate": 4.474789915966387e-05,
+      "loss": 0.0882,
+      "step": 2175
+    },
+    {
+      "epoch": 10.95,
+      "learning_rate": 4.369747899159664e-05,
+      "loss": 0.0806,
+      "step": 2180
+    },
+    {
+      "epoch": 10.98,
+      "learning_rate": 4.2647058823529415e-05,
+      "loss": 0.1775,
+      "step": 2185
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.9665252239509665,
+      "eval_loss": 0.08594423532485962,
+      "eval_runtime": 71.6846,
+      "eval_samples_per_second": 29.588,
+      "eval_steps_per_second": 1.855,
+      "step": 2189
+    },
+    {
+      "epoch": 11.01,
+      "learning_rate": 4.159663865546219e-05,
+      "loss": 0.0979,
+      "step": 2190
+    },
+    {
+      "epoch": 11.03,
+      "learning_rate": 4.054621848739496e-05,
+      "loss": 0.1205,
+      "step": 2195
+    },
+    {
+      "epoch": 11.06,
+      "learning_rate": 3.949579831932773e-05,
+      "loss": 0.0242,
+      "step": 2200
+    },
+    {
+      "epoch": 11.08,
+      "learning_rate": 3.844537815126051e-05,
+      "loss": 0.0271,
+      "step": 2205
+    },
+    {
+      "epoch": 11.11,
+      "learning_rate": 3.739495798319328e-05,
+      "loss": 0.1402,
+      "step": 2210
+    },
+    {
+      "epoch": 11.13,
+      "learning_rate": 3.634453781512605e-05,
+      "loss": 0.1164,
+      "step": 2215
+    },
+    {
+      "epoch": 11.16,
+      "learning_rate": 3.529411764705883e-05,
+      "loss": 0.1977,
+      "step": 2220
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 3.42436974789916e-05,
+      "loss": 0.2848,
+      "step": 2225
+    },
+    {
+      "epoch": 11.21,
+      "learning_rate": 3.319327731092437e-05,
+      "loss": 0.1909,
+      "step": 2230
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 3.2142857142857144e-05,
+      "loss": 0.1461,
+      "step": 2235
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 3.1092436974789916e-05,
+      "loss": 0.1888,
+      "step": 2240
+    },
+    {
+      "epoch": 11.28,
+      "learning_rate": 3.004201680672269e-05,
+      "loss": 0.0977,
+      "step": 2245
+    },
+    {
+      "epoch": 11.31,
+      "learning_rate": 2.8991596638655467e-05,
+      "loss": 0.1312,
+      "step": 2250
+    },
+    {
+      "epoch": 11.33,
+      "learning_rate": 2.7941176470588236e-05,
+      "loss": 0.0771,
+      "step": 2255
+    },
+    {
+      "epoch": 11.36,
+      "learning_rate": 2.689075630252101e-05,
+      "loss": 0.1002,
+      "step": 2260
+    },
+    {
+      "epoch": 11.38,
+      "learning_rate": 2.5840336134453784e-05,
+      "loss": 0.1001,
+      "step": 2265
+    },
+    {
+      "epoch": 11.41,
+      "learning_rate": 2.4789915966386556e-05,
+      "loss": 0.0841,
+      "step": 2270
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 2.373949579831933e-05,
+      "loss": 0.089,
+      "step": 2275
+    },
+    {
+      "epoch": 11.46,
+      "learning_rate": 2.26890756302521e-05,
+      "loss": 0.1083,
+      "step": 2280
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 2.1638655462184876e-05,
+      "loss": 0.0431,
+      "step": 2285
+    },
+    {
+      "epoch": 11.51,
+      "learning_rate": 2.058823529411765e-05,
+      "loss": 0.0761,
+      "step": 2290
+    },
+    {
+      "epoch": 11.53,
+      "learning_rate": 1.953781512605042e-05,
+      "loss": 0.0842,
+      "step": 2295
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 1.8487394957983196e-05,
+      "loss": 0.0914,
+      "step": 2300
+    },
+    {
+      "epoch": 11.58,
+      "learning_rate": 1.7436974789915965e-05,
+      "loss": 0.091,
+      "step": 2305
+    },
+    {
+      "epoch": 11.61,
+      "learning_rate": 1.638655462184874e-05,
+      "loss": 0.0254,
+      "step": 2310
+    },
+    {
+      "epoch": 11.63,
+      "learning_rate": 1.5336134453781513e-05,
+      "loss": 0.0794,
+      "step": 2315
+    },
+    {
+      "epoch": 11.66,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.0469,
+      "step": 2320
+    },
+    {
+      "epoch": 11.68,
+      "learning_rate": 1.323529411764706e-05,
+      "loss": 0.0586,
+      "step": 2325
+    },
+    {
+      "epoch": 11.71,
+      "learning_rate": 1.2184873949579832e-05,
+      "loss": 0.051,
+      "step": 2330
+    },
+    {
+      "epoch": 11.73,
+      "learning_rate": 1.1134453781512606e-05,
+      "loss": 0.0594,
+      "step": 2335
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 1.008403361344538e-05,
+      "loss": 0.0456,
+      "step": 2340
+    },
+    {
+      "epoch": 11.78,
+      "learning_rate": 9.033613445378152e-06,
+      "loss": 0.0164,
+      "step": 2345
+    },
+    {
+      "epoch": 11.81,
+      "learning_rate": 7.983193277310924e-06,
+      "loss": 0.042,
+      "step": 2350
+    },
+    {
+      "epoch": 11.83,
+      "learning_rate": 6.932773109243698e-06,
+      "loss": 0.0351,
+      "step": 2355
+    },
+    {
+      "epoch": 11.86,
+      "learning_rate": 5.882352941176471e-06,
+      "loss": 0.0654,
+      "step": 2360
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 4.831932773109244e-06,
+      "loss": 0.027,
+      "step": 2365
+    },
+    {
+      "epoch": 11.91,
+      "learning_rate": 3.7815126050420167e-06,
+      "loss": 0.0912,
+      "step": 2370
+    },
+    {
+      "epoch": 11.93,
+      "learning_rate": 2.73109243697479e-06,
+      "loss": 0.0446,
+      "step": 2375
+    },
+    {
+      "epoch": 11.96,
+      "learning_rate": 1.680672268907563e-06,
+      "loss": 0.0666,
+      "step": 2380
+    },
+    {
+      "epoch": 11.98,
+      "learning_rate": 6.302521008403362e-07,
+      "loss": 0.0042,
+      "step": 2385
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.9900990099009901,
+      "eval_loss": 0.03527973219752312,
+      "eval_runtime": 71.6307,
+      "eval_samples_per_second": 29.61,
+      "eval_steps_per_second": 1.857,
+      "step": 2388
+    },
+    {
+      "epoch": 12.0,
+      "step": 2388,
+      "total_flos": 1.3545775098737345e+19,
+      "train_loss": 0.23949940906840703,
+      "train_runtime": 11760.3071,
+      "train_samples_per_second": 6.493,
+      "train_steps_per_second": 0.203
+    }
+  ],
+  "max_steps": 2388,
+  "num_train_epochs": 12,
+  "total_flos": 1.3545775098737345e+19,
+  "trial_name": null,
+  "trial_params": null
+}