Upload 10 files

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e61ea2e45edfeec896776af115dabcc98703ee8cc296de762c6e7c1a1defc783
+oid sha256:5e6bf3bde909bd33d0b254dc94556f1eff9f965c1d44cbf806d02a903084b53c
 size 4473979569

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f8c1339fd3044159d03f5d1792c685c645aa011d4a846dd27b56d22cb71fb26
+oid sha256:2a14d36d54c719d270aa81ebfee4a5639b67b9d1e6c3c20f8da870ef41bf510d
 size 2236965925

trainer_state.json

@@ -10,907 +10,907 @@
     {
       "epoch": 0.0,
       "learning_rate": 4.9971428571428576e-05,
-      "loss": 13.0691,
+      "loss": 20.0833,
       "step": 1
     },
     {
       "epoch": 0.02,
       "learning_rate": 4.9e-05,
-      "loss": 14.048,
+      "loss": 7.9373,
       "step": 35
     },
     {
       "epoch": 0.02,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.6666666666666666,
-      "eval_loss": 7.880705833435059,
-      "eval_precision": 0.5,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.7039,
-      "eval_samples_per_second": 37.173,
-      "eval_steps_per_second": 4.647,
+      "eval_accuracy": 0.614,
+      "eval_f1": 0.7190684133915575,
+      "eval_loss": 2.977444648742676,
+      "eval_precision": 0.5652173913043478,
+      "eval_recall": 0.988,
+      "eval_runtime": 80.901,
+      "eval_samples_per_second": 37.082,
+      "eval_steps_per_second": 4.635,
       "step": 35
     },
     {
       "epoch": 0.04,
       "learning_rate": 4.8e-05,
-      "loss": 2.2656,
+      "loss": 1.0358,
       "step": 70
     },
     {
       "epoch": 0.04,
-      "eval_accuracy": 0.5016666666666667,
-      "eval_f1": 0.6674082313681868,
-      "eval_loss": 3.2190163135528564,
-      "eval_precision": 0.5008347245409015,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.4061,
-      "eval_samples_per_second": 37.311,
-      "eval_steps_per_second": 4.664,
+      "eval_accuracy": 0.6373333333333333,
+      "eval_f1": 0.7332025502697401,
+      "eval_loss": 3.09529972076416,
+      "eval_precision": 0.5799069045771916,
+      "eval_recall": 0.9966666666666667,
+      "eval_runtime": 82.2367,
+      "eval_samples_per_second": 36.48,
+      "eval_steps_per_second": 4.56,
       "step": 70
     },
     {
       "epoch": 0.06,
       "learning_rate": 4.7e-05,
-      "loss": 2.1548,
+      "loss": 1.8177,
       "step": 105
     },
     {
       "epoch": 0.06,
-      "eval_accuracy": 0.6906666666666667,
-      "eval_f1": 0.7371104815864024,
-      "eval_loss": 0.5703864097595215,
-      "eval_precision": 0.6408866995073892,
-      "eval_recall": 0.8673333333333333,
-      "eval_runtime": 80.8676,
-      "eval_samples_per_second": 37.098,
-      "eval_steps_per_second": 4.637,
+      "eval_accuracy": 0.8706666666666667,
+      "eval_f1": 0.8636683063949402,
+      "eval_loss": 0.640018105506897,
+      "eval_precision": 0.913075780089153,
+      "eval_recall": 0.8193333333333334,
+      "eval_runtime": 81.9619,
+      "eval_samples_per_second": 36.602,
+      "eval_steps_per_second": 4.575,
       "step": 105
     },
     {
       "epoch": 0.08,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 1.3819,
+      "loss": 0.7211,
       "step": 140
     },
     {
       "epoch": 0.08,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.0,
-      "eval_loss": 0.9337035417556763,
-      "eval_precision": 0.0,
-      "eval_recall": 0.0,
-      "eval_runtime": 80.5811,
-      "eval_samples_per_second": 37.23,
-      "eval_steps_per_second": 4.654,
+      "eval_accuracy": 0.821,
+      "eval_f1": 0.811777076761304,
+      "eval_loss": 1.0340842008590698,
+      "eval_precision": 0.8558758314855875,
+      "eval_recall": 0.772,
+      "eval_runtime": 82.2896,
+      "eval_samples_per_second": 36.457,
+      "eval_steps_per_second": 4.557,
       "step": 140
     },
     {
       "epoch": 0.1,
       "learning_rate": 4.5e-05,
-      "loss": 1.1309,
+      "loss": 1.4119,
       "step": 175
     },
     {
       "epoch": 0.1,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.6666666666666666,
-      "eval_loss": 2.4655497074127197,
-      "eval_precision": 0.5,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.4828,
-      "eval_samples_per_second": 37.275,
-      "eval_steps_per_second": 4.659,
+      "eval_accuracy": 0.751,
+      "eval_f1": 0.6828025477707006,
+      "eval_loss": 0.7746813893318176,
+      "eval_precision": 0.9403508771929825,
+      "eval_recall": 0.536,
+      "eval_runtime": 82.2817,
+      "eval_samples_per_second": 36.46,
+      "eval_steps_per_second": 4.558,
       "step": 175
     },
     {
       "epoch": 0.12,
       "learning_rate": 4.4000000000000006e-05,
-      "loss": 0.9397,
+      "loss": 0.8471,
       "step": 210
     },
     {
       "epoch": 0.12,
-      "eval_accuracy": 0.6686666666666666,
-      "eval_f1": 0.7317862924986509,
-      "eval_loss": 0.6442784667015076,
-      "eval_precision": 0.614687216681777,
-      "eval_recall": 0.904,
-      "eval_runtime": 80.537,
-      "eval_samples_per_second": 37.25,
-      "eval_steps_per_second": 4.656,
+      "eval_accuracy": 0.854,
+      "eval_f1": 0.8643122676579925,
+      "eval_loss": 0.6877764463424683,
+      "eval_precision": 0.8072916666666666,
+      "eval_recall": 0.93,
+      "eval_runtime": 81.2562,
+      "eval_samples_per_second": 36.92,
+      "eval_steps_per_second": 4.615,
       "step": 210
     },
     {
       "epoch": 0.14,
       "learning_rate": 4.3e-05,
-      "loss": 0.8193,
+      "loss": 1.3883,
       "step": 245
     },
     {
       "epoch": 0.14,
-      "eval_accuracy": 0.6676666666666666,
-      "eval_f1": 0.7155492154065621,
-      "eval_loss": 1.0818308591842651,
-      "eval_precision": 0.6254364089775561,
-      "eval_recall": 0.836,
-      "eval_runtime": 81.1642,
-      "eval_samples_per_second": 36.962,
-      "eval_steps_per_second": 4.62,
+      "eval_accuracy": 0.7186666666666667,
+      "eval_f1": 0.6114180478821363,
+      "eval_loss": 2.117485284805298,
+      "eval_precision": 0.9880952380952381,
+      "eval_recall": 0.44266666666666665,
+      "eval_runtime": 81.2535,
+      "eval_samples_per_second": 36.921,
+      "eval_steps_per_second": 4.615,
       "step": 245
     },
     {
       "epoch": 0.16,
       "learning_rate": 4.2e-05,
-      "loss": 0.6235,
+      "loss": 0.9955,
       "step": 280
     },
     {
       "epoch": 0.16,
-      "eval_accuracy": 0.5413333333333333,
-      "eval_f1": 0.685413808870599,
-      "eval_loss": 1.136587381362915,
-      "eval_precision": 0.5215727209464162,
-      "eval_recall": 0.9993333333333333,
-      "eval_runtime": 81.1897,
-      "eval_samples_per_second": 36.951,
-      "eval_steps_per_second": 4.619,
+      "eval_accuracy": 0.853,
+      "eval_f1": 0.8333962976955043,
+      "eval_loss": 0.9773063063621521,
+      "eval_precision": 0.9616390584132519,
+      "eval_recall": 0.7353333333333333,
+      "eval_runtime": 82.2707,
+      "eval_samples_per_second": 36.465,
+      "eval_steps_per_second": 4.558,
       "step": 280
     },
     {
       "epoch": 0.18,
       "learning_rate": 4.1e-05,
-      "loss": 0.6828,
+      "loss": 0.9704,
       "step": 315
     },
     {
       "epoch": 0.18,
-      "eval_accuracy": 0.7913333333333333,
-      "eval_f1": 0.7481898632341111,
-      "eval_loss": 0.5717737674713135,
-      "eval_precision": 0.9432048681541582,
-      "eval_recall": 0.62,
-      "eval_runtime": 80.6162,
-      "eval_samples_per_second": 37.213,
-      "eval_steps_per_second": 4.652,
+      "eval_accuracy": 0.8093333333333333,
+      "eval_f1": 0.8361970217640321,
+      "eval_loss": 0.7936842441558838,
+      "eval_precision": 0.7329317269076305,
+      "eval_recall": 0.9733333333333334,
+      "eval_runtime": 81.186,
+      "eval_samples_per_second": 36.952,
+      "eval_steps_per_second": 4.619,
       "step": 315
     },
     {
       "epoch": 0.2,
       "learning_rate": 4e-05,
-      "loss": 0.5375,
+      "loss": 1.3156,
       "step": 350
     },
     {
       "epoch": 0.2,
-      "eval_accuracy": 0.6203333333333333,
-      "eval_f1": 0.7234765719834911,
-      "eval_loss": 1.3898557424545288,
-      "eval_precision": 0.5689194348988164,
-      "eval_recall": 0.9933333333333333,
-      "eval_runtime": 80.8767,
-      "eval_samples_per_second": 37.094,
-      "eval_steps_per_second": 4.637,
+      "eval_accuracy": 0.7793333333333333,
+      "eval_f1": 0.8049499116087213,
+      "eval_loss": 1.306660771369934,
+      "eval_precision": 0.7212249208025343,
+      "eval_recall": 0.9106666666666666,
+      "eval_runtime": 81.1695,
+      "eval_samples_per_second": 36.96,
+      "eval_steps_per_second": 4.62,
       "step": 350
     },
     {
       "epoch": 0.22,
       "learning_rate": 3.9000000000000006e-05,
-      "loss": 0.8008,
+      "loss": 0.9042,
       "step": 385
     },
     {
       "epoch": 0.22,
-      "eval_accuracy": 0.6446666666666667,
-      "eval_f1": 0.7352210630899155,
-      "eval_loss": 1.3506994247436523,
-      "eval_precision": 0.5859065716547902,
-      "eval_recall": 0.9866666666666667,
-      "eval_runtime": 80.9272,
-      "eval_samples_per_second": 37.07,
-      "eval_steps_per_second": 4.634,
+      "eval_accuracy": 0.7506666666666667,
+      "eval_f1": 0.6978998384491114,
+      "eval_loss": 0.8157205581665039,
+      "eval_precision": 0.8852459016393442,
+      "eval_recall": 0.576,
+      "eval_runtime": 81.9836,
+      "eval_samples_per_second": 36.593,
+      "eval_steps_per_second": 4.574,
       "step": 385
     },
     {
       "epoch": 0.24,
       "learning_rate": 3.8e-05,
-      "loss": 0.5408,
+      "loss": 0.6122,
       "step": 420
     },
     {
       "epoch": 0.24,
-      "eval_accuracy": 0.5,
-      "eval_f1": 0.6666666666666666,
-      "eval_loss": 5.448528289794922,
-      "eval_precision": 0.5,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.3832,
-      "eval_samples_per_second": 37.321,
-      "eval_steps_per_second": 4.665,
+      "eval_accuracy": 0.861,
+      "eval_f1": 0.8741322064594024,
+      "eval_loss": 0.6521239280700684,
+      "eval_precision": 0.7986762272476559,
+      "eval_recall": 0.9653333333333334,
+      "eval_runtime": 81.1912,
+      "eval_samples_per_second": 36.95,
+      "eval_steps_per_second": 4.619,
       "step": 420
     },
     {
       "epoch": 0.26,
       "learning_rate": 3.7e-05,
-      "loss": 1.5679,
+      "loss": 0.9266,
       "step": 455
     },
     {
       "epoch": 0.26,
-      "eval_accuracy": 0.5303333333333333,
-      "eval_f1": 0.6804264005443411,
-      "eval_loss": 2.9407122135162354,
-      "eval_precision": 0.5156411137848058,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.4322,
-      "eval_samples_per_second": 37.298,
-      "eval_steps_per_second": 4.662,
+      "eval_accuracy": 0.7576666666666667,
+      "eval_f1": 0.8008764721993973,
+      "eval_loss": 1.0884476900100708,
+      "eval_precision": 0.6796838679683868,
+      "eval_recall": 0.9746666666666667,
+      "eval_runtime": 81.1743,
+      "eval_samples_per_second": 36.958,
+      "eval_steps_per_second": 4.62,
       "step": 455
     },
     {
       "epoch": 0.28,
       "learning_rate": 3.6e-05,
-      "loss": 0.7113,
+      "loss": 0.5418,
       "step": 490
     },
     {
       "epoch": 0.28,
-      "eval_accuracy": 0.5003333333333333,
-      "eval_f1": 0.6668148477439431,
-      "eval_loss": 4.913637638092041,
-      "eval_precision": 0.5001667222407469,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.9405,
-      "eval_samples_per_second": 37.064,
-      "eval_steps_per_second": 4.633,
+      "eval_accuracy": 0.8626666666666667,
+      "eval_f1": 0.8731527093596059,
+      "eval_loss": 0.594514012336731,
+      "eval_precision": 0.8112128146453089,
+      "eval_recall": 0.9453333333333334,
+      "eval_runtime": 81.6996,
+      "eval_samples_per_second": 36.72,
+      "eval_steps_per_second": 4.59,
       "step": 490
     },
     {
       "epoch": 0.3,
       "learning_rate": 3.5e-05,
-      "loss": 0.5356,
+      "loss": 0.4564,
       "step": 525
     },
     {
       "epoch": 0.3,
-      "eval_accuracy": 0.6223333333333333,
-      "eval_f1": 0.724933236222384,
-      "eval_loss": 1.4003673791885376,
-      "eval_precision": 0.5700649102710958,
-      "eval_recall": 0.9953333333333333,
-      "eval_runtime": 80.635,
-      "eval_samples_per_second": 37.205,
-      "eval_steps_per_second": 4.651,
+      "eval_accuracy": 0.9256666666666666,
+      "eval_f1": 0.9236039739636861,
+      "eval_loss": 0.2299291044473648,
+      "eval_precision": 0.9499647639182522,
+      "eval_recall": 0.8986666666666666,
+      "eval_runtime": 81.1873,
+      "eval_samples_per_second": 36.952,
+      "eval_steps_per_second": 4.619,
       "step": 525
     },
     {
       "epoch": 0.32,
       "learning_rate": 3.4000000000000007e-05,
-      "loss": 0.4847,
+      "loss": 0.427,
       "step": 560
     },
     {
       "epoch": 0.32,
-      "eval_accuracy": 0.5016666666666667,
-      "eval_f1": 0.6674082313681868,
-      "eval_loss": 6.015239715576172,
-      "eval_precision": 0.5008347245409015,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.7702,
-      "eval_samples_per_second": 37.142,
-      "eval_steps_per_second": 4.643,
+      "eval_accuracy": 0.9156666666666666,
+      "eval_f1": 0.9154694286668894,
+      "eval_loss": 0.36590951681137085,
+      "eval_precision": 0.9176155391828533,
+      "eval_recall": 0.9133333333333333,
+      "eval_runtime": 81.3624,
+      "eval_samples_per_second": 36.872,
+      "eval_steps_per_second": 4.609,
       "step": 560
     },
     {
       "epoch": 0.34,
       "learning_rate": 3.3e-05,
-      "loss": 0.319,
+      "loss": 0.7026,
       "step": 595
     },
     {
       "epoch": 0.34,
-      "eval_accuracy": 0.8496666666666667,
-      "eval_f1": 0.8543752018082015,
-      "eval_loss": 0.6166228652000427,
-      "eval_precision": 0.828428303068253,
-      "eval_recall": 0.882,
-      "eval_runtime": 80.8211,
-      "eval_samples_per_second": 37.119,
-      "eval_steps_per_second": 4.64,
+      "eval_accuracy": 0.9166666666666666,
+      "eval_f1": 0.9150815217391304,
+      "eval_loss": 0.509623646736145,
+      "eval_precision": 0.932825484764543,
+      "eval_recall": 0.898,
+      "eval_runtime": 81.284,
+      "eval_samples_per_second": 36.908,
+      "eval_steps_per_second": 4.613,
       "step": 595
     },
     {
       "epoch": 0.36,
       "learning_rate": 3.2000000000000005e-05,
-      "loss": 0.822,
+      "loss": 0.7019,
       "step": 630
     },
     {
       "epoch": 0.36,
-      "eval_accuracy": 0.8383333333333334,
-      "eval_f1": 0.820967146548542,
-      "eval_loss": 0.41078850626945496,
-      "eval_precision": 0.9197684036393714,
-      "eval_recall": 0.7413333333333333,
-      "eval_runtime": 80.9973,
-      "eval_samples_per_second": 37.038,
-      "eval_steps_per_second": 4.63,
+      "eval_accuracy": 0.9066666666666666,
+      "eval_f1": 0.9068529607451763,
+      "eval_loss": 0.4853467643260956,
+      "eval_precision": 0.9050464807436919,
+      "eval_recall": 0.9086666666666666,
+      "eval_runtime": 81.2065,
+      "eval_samples_per_second": 36.943,
+      "eval_steps_per_second": 4.618,
       "step": 630
     },
     {
       "epoch": 0.38,
       "learning_rate": 3.1e-05,
-      "loss": 0.4746,
+      "loss": 0.5116,
       "step": 665
     },
     {
       "epoch": 0.38,
-      "eval_accuracy": 0.8246666666666667,
-      "eval_f1": 0.830213040671401,
-      "eval_loss": 0.3808746337890625,
-      "eval_precision": 0.804755944931164,
-      "eval_recall": 0.8573333333333333,
-      "eval_runtime": 80.4252,
-      "eval_samples_per_second": 37.302,
-      "eval_steps_per_second": 4.663,
+      "eval_accuracy": 0.9003333333333333,
+      "eval_f1": 0.9047467346288627,
+      "eval_loss": 0.4890832006931305,
+      "eval_precision": 0.8663819402074435,
+      "eval_recall": 0.9466666666666667,
+      "eval_runtime": 81.3158,
+      "eval_samples_per_second": 36.893,
+      "eval_steps_per_second": 4.612,
       "step": 665
     },
     {
       "epoch": 0.4,
       "learning_rate": 3e-05,
-      "loss": 0.2913,
+      "loss": 0.4625,
       "step": 700
     },
     {
       "epoch": 0.4,
-      "eval_accuracy": 0.747,
-      "eval_f1": 0.7941415785191213,
-      "eval_loss": 0.7681821584701538,
-      "eval_precision": 0.6694101508916324,
-      "eval_recall": 0.976,
-      "eval_runtime": 80.4193,
-      "eval_samples_per_second": 37.304,
-      "eval_steps_per_second": 4.663,
+      "eval_accuracy": 0.9276666666666666,
+      "eval_f1": 0.92340275326509,
+      "eval_loss": 0.391652911901474,
+      "eval_precision": 0.981245311327832,
+      "eval_recall": 0.872,
+      "eval_runtime": 81.251,
+      "eval_samples_per_second": 36.923,
+      "eval_steps_per_second": 4.615,
       "step": 700
     },
     {
       "epoch": 0.42,
       "learning_rate": 2.9e-05,
-      "loss": 0.3153,
+      "loss": 0.455,
       "step": 735
     },
     {
       "epoch": 0.42,
-      "eval_accuracy": 0.8356666666666667,
-      "eval_f1": 0.8523510032943996,
-      "eval_loss": 0.4611086845397949,
-      "eval_precision": 0.7737901033170201,
-      "eval_recall": 0.9486666666666667,
-      "eval_runtime": 80.4421,
-      "eval_samples_per_second": 37.294,
-      "eval_steps_per_second": 4.662,
+      "eval_accuracy": 0.89,
+      "eval_f1": 0.9003623188405797,
+      "eval_loss": 0.5772602558135986,
+      "eval_precision": 0.8228476821192053,
+      "eval_recall": 0.994,
+      "eval_runtime": 81.2335,
+      "eval_samples_per_second": 36.931,
+      "eval_steps_per_second": 4.616,
       "step": 735
     },
     {
       "epoch": 0.44,
       "learning_rate": 2.8000000000000003e-05,
-      "loss": 0.4953,
+      "loss": 0.3816,
       "step": 770
     },
     {
       "epoch": 0.44,
-      "eval_accuracy": 0.5146666666666667,
-      "eval_f1": 0.6732495511669659,
-      "eval_loss": 1.9276390075683594,
-      "eval_precision": 0.5074424898511503,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.6277,
-      "eval_samples_per_second": 37.208,
-      "eval_steps_per_second": 4.651,
+      "eval_accuracy": 0.929,
+      "eval_f1": 0.9305510270622758,
+      "eval_loss": 0.32315143942832947,
+      "eval_precision": 0.9106573069559668,
+      "eval_recall": 0.9513333333333334,
+      "eval_runtime": 81.1544,
+      "eval_samples_per_second": 36.967,
+      "eval_steps_per_second": 4.621,
       "step": 770
     },
     {
       "epoch": 0.46,
       "learning_rate": 2.7000000000000002e-05,
-      "loss": 0.8238,
+      "loss": 0.3276,
       "step": 805
     },
     {
       "epoch": 0.46,
-      "eval_accuracy": 0.8233333333333334,
-      "eval_f1": 0.789515488482923,
-      "eval_loss": 0.758985698223114,
-      "eval_precision": 0.9764243614931237,
-      "eval_recall": 0.6626666666666666,
-      "eval_runtime": 80.4659,
-      "eval_samples_per_second": 37.283,
-      "eval_steps_per_second": 4.66,
+      "eval_accuracy": 0.894,
+      "eval_f1": 0.9034608378870674,
+      "eval_loss": 0.8502470850944519,
+      "eval_precision": 0.8294314381270903,
+      "eval_recall": 0.992,
+      "eval_runtime": 81.1738,
+      "eval_samples_per_second": 36.958,
+      "eval_steps_per_second": 4.62,
       "step": 805
     },
     {
       "epoch": 0.48,
       "learning_rate": 2.6000000000000002e-05,
-      "loss": 0.6315,
+      "loss": 0.8028,
       "step": 840
     },
     {
       "epoch": 0.48,
-      "eval_accuracy": 0.5013333333333333,
-      "eval_f1": 0.6672597864768683,
-      "eval_loss": 5.900465965270996,
-      "eval_precision": 0.5006675567423231,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.4747,
-      "eval_samples_per_second": 37.279,
-      "eval_steps_per_second": 4.66,
+      "eval_accuracy": 0.8066666666666666,
+      "eval_f1": 0.761904761904762,
+      "eval_loss": 1.2200298309326172,
+      "eval_precision": 0.9914529914529915,
+      "eval_recall": 0.6186666666666667,
+      "eval_runtime": 81.3151,
+      "eval_samples_per_second": 36.894,
+      "eval_steps_per_second": 4.612,
       "step": 840
     },
     {
       "epoch": 0.5,
       "learning_rate": 2.5e-05,
-      "loss": 0.4659,
+      "loss": 0.4308,
       "step": 875
     },
     {
       "epoch": 0.5,
-      "eval_accuracy": 0.879,
-      "eval_f1": 0.875214850464077,
-      "eval_loss": 0.2947004735469818,
-      "eval_precision": 0.9034776437189496,
-      "eval_recall": 0.8486666666666667,
-      "eval_runtime": 80.8565,
-      "eval_samples_per_second": 37.103,
-      "eval_steps_per_second": 4.638,
+      "eval_accuracy": 0.9213333333333333,
+      "eval_f1": 0.9263880224578914,
+      "eval_loss": 0.36110782623291016,
+      "eval_precision": 0.8704572098475967,
+      "eval_recall": 0.99,
+      "eval_runtime": 81.936,
+      "eval_samples_per_second": 36.614,
+      "eval_steps_per_second": 4.577,
       "step": 875
     },
     {
       "epoch": 0.52,
       "learning_rate": 2.4e-05,
-      "loss": 0.2773,
+      "loss": 0.3567,
       "step": 910
     },
     {
       "epoch": 0.52,
-      "eval_accuracy": 0.5133333333333333,
-      "eval_f1": 0.672645739910314,
-      "eval_loss": 3.5455527305603027,
-      "eval_precision": 0.5067567567567568,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.4994,
-      "eval_samples_per_second": 37.267,
-      "eval_steps_per_second": 4.658,
+      "eval_accuracy": 0.908,
+      "eval_f1": 0.9000724112961622,
+      "eval_loss": 0.6623110771179199,
+      "eval_precision": 0.9849445324881141,
+      "eval_recall": 0.8286666666666667,
+      "eval_runtime": 81.4083,
+      "eval_samples_per_second": 36.851,
+      "eval_steps_per_second": 4.606,
       "step": 910
     },
     {
       "epoch": 0.54,
       "learning_rate": 2.3000000000000003e-05,
-      "loss": 0.371,
+      "loss": 0.4286,
       "step": 945
     },
     {
       "epoch": 0.54,
-      "eval_accuracy": 0.5043333333333333,
-      "eval_f1": 0.6685981724983284,
-      "eval_loss": 3.9809491634368896,
-      "eval_precision": 0.5021760964178105,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.5845,
-      "eval_samples_per_second": 37.228,
-      "eval_steps_per_second": 4.654,
+      "eval_accuracy": 0.9216666666666666,
+      "eval_f1": 0.9264475743348982,
+      "eval_loss": 0.3584176301956177,
+      "eval_precision": 0.8731563421828908,
+      "eval_recall": 0.9866666666666667,
+      "eval_runtime": 81.0909,
+      "eval_samples_per_second": 36.996,
+      "eval_steps_per_second": 4.624,
       "step": 945
     },
     {
       "epoch": 0.56,
       "learning_rate": 2.2000000000000003e-05,
-      "loss": 0.4968,
+      "loss": 0.3559,
       "step": 980
     },
     {
       "epoch": 0.56,
-      "eval_accuracy": 0.713,
-      "eval_f1": 0.7757228444907528,
-      "eval_loss": 1.2047169208526611,
-      "eval_precision": 0.6365968362548098,
-      "eval_recall": 0.9926666666666667,
-      "eval_runtime": 80.9051,
-      "eval_samples_per_second": 37.08,
-      "eval_steps_per_second": 4.635,
+      "eval_accuracy": 0.944,
+      "eval_f1": 0.945736434108527,
+      "eval_loss": 0.21195654571056366,
+      "eval_precision": 0.9172932330827067,
+      "eval_recall": 0.976,
+      "eval_runtime": 81.1261,
+      "eval_samples_per_second": 36.979,
+      "eval_steps_per_second": 4.622,
       "step": 980
     },
     {
       "epoch": 0.58,
       "learning_rate": 2.1e-05,
-      "loss": 0.5577,
+      "loss": 0.3043,
       "step": 1015
     },
     {
       "epoch": 0.58,
-      "eval_accuracy": 0.9123333333333333,
-      "eval_f1": 0.9118337244384848,
-      "eval_loss": 0.24421826004981995,
-      "eval_precision": 0.9170600134861767,
-      "eval_recall": 0.9066666666666666,
-      "eval_runtime": 80.8404,
-      "eval_samples_per_second": 37.11,
-      "eval_steps_per_second": 4.639,
+      "eval_accuracy": 0.923,
+      "eval_f1": 0.9262216544235068,
+      "eval_loss": 0.23412927985191345,
+      "eval_precision": 0.8890251379521765,
+      "eval_recall": 0.9666666666666667,
+      "eval_runtime": 81.1527,
+      "eval_samples_per_second": 36.967,
+      "eval_steps_per_second": 4.621,
       "step": 1015
     },
     {
       "epoch": 0.6,
       "learning_rate": 2e-05,
-      "loss": 0.2212,
+      "loss": 0.207,
       "step": 1050
     },
     {
       "epoch": 0.6,
-      "eval_accuracy": 0.8366666666666667,
-      "eval_f1": 0.8556275780789628,
-      "eval_loss": 0.44433295726776123,
-      "eval_precision": 0.7666314677930306,
-      "eval_recall": 0.968,
-      "eval_runtime": 80.4427,
-      "eval_samples_per_second": 37.294,
-      "eval_steps_per_second": 4.662,
+      "eval_accuracy": 0.9186666666666666,
+      "eval_f1": 0.9122302158273381,
+      "eval_loss": 0.4979284703731537,
+      "eval_precision": 0.990625,
+      "eval_recall": 0.8453333333333334,
+      "eval_runtime": 81.616,
+      "eval_samples_per_second": 36.757,
+      "eval_steps_per_second": 4.595,
       "step": 1050
     },
     {
       "epoch": 0.62,
       "learning_rate": 1.9e-05,
-      "loss": 0.2668,
+      "loss": 0.4594,
       "step": 1085
     },
     {
       "epoch": 0.62,
-      "eval_accuracy": 0.56,
-      "eval_f1": 0.6943029180176007,
-      "eval_loss": 2.2121241092681885,
-      "eval_precision": 0.5319375443577005,
-      "eval_recall": 0.9993333333333333,
-      "eval_runtime": 80.772,
-      "eval_samples_per_second": 37.142,
-      "eval_steps_per_second": 4.643,
+      "eval_accuracy": 0.9316666666666666,
+      "eval_f1": 0.9276385457112601,
+      "eval_loss": 0.33773142099380493,
+      "eval_precision": 0.9857464366091523,
+      "eval_recall": 0.876,
+      "eval_runtime": 81.8837,
+      "eval_samples_per_second": 36.637,
+      "eval_steps_per_second": 4.58,
       "step": 1085
     },
     {
       "epoch": 0.64,
       "learning_rate": 1.8e-05,
-      "loss": 0.2392,
+      "loss": 0.3886,
       "step": 1120
     },
     {
       "epoch": 0.64,
-      "eval_accuracy": 0.617,
-      "eval_f1": 0.7223967141821697,
-      "eval_loss": 1.7039114236831665,
-      "eval_precision": 0.5665024630541872,
-      "eval_recall": 0.9966666666666667,
-      "eval_runtime": 81.2102,
-      "eval_samples_per_second": 36.941,
-      "eval_steps_per_second": 4.618,
+      "eval_accuracy": 0.952,
+      "eval_f1": 0.9511533242876526,
+      "eval_loss": 0.21586619317531586,
+      "eval_precision": 0.9682320441988951,
+      "eval_recall": 0.9346666666666666,
+      "eval_runtime": 81.9925,
+      "eval_samples_per_second": 36.589,
+      "eval_steps_per_second": 4.574,
       "step": 1120
     },
     {
       "epoch": 0.66,
       "learning_rate": 1.7000000000000003e-05,
-      "loss": 0.4237,
+      "loss": 0.3025,
       "step": 1155
     },
     {
       "epoch": 0.66,
-      "eval_accuracy": 0.5113333333333333,
-      "eval_f1": 0.6717420510523958,
-      "eval_loss": 3.3159143924713135,
-      "eval_precision": 0.5057316250842886,
-      "eval_recall": 1.0,
-      "eval_runtime": 80.4636,
-      "eval_samples_per_second": 37.284,
-      "eval_steps_per_second": 4.66,
+      "eval_accuracy": 0.9516666666666667,
+      "eval_f1": 0.9511620074099023,
+      "eval_loss": 0.1657862514257431,
+      "eval_precision": 0.9611980939414567,
+      "eval_recall": 0.9413333333333334,
+      "eval_runtime": 81.6092,
+      "eval_samples_per_second": 36.761,
+      "eval_steps_per_second": 4.595,
       "step": 1155
     },
     {
       "epoch": 0.68,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.4353,
+      "loss": 0.3561,
       "step": 1190
     },
     {
       "epoch": 0.68,
-      "eval_accuracy": 0.82,
-      "eval_f1": 0.8432055749128919,
-      "eval_loss": 0.5704355239868164,
-      "eval_precision": 0.7469135802469136,
-      "eval_recall": 0.968,
-      "eval_runtime": 80.5065,
-      "eval_samples_per_second": 37.264,
-      "eval_steps_per_second": 4.658,
+      "eval_accuracy": 0.926,
+      "eval_f1": 0.9304075235109718,
+      "eval_loss": 0.2859382629394531,
+      "eval_precision": 0.8781065088757396,
+      "eval_recall": 0.9893333333333333,
+      "eval_runtime": 81.4491,
+      "eval_samples_per_second": 36.833,
+      "eval_steps_per_second": 4.604,
       "step": 1190
     },
     {
       "epoch": 0.7,
       "learning_rate": 1.5e-05,
-      "loss": 0.3109,
+      "loss": 0.2377,
       "step": 1225
     },
     {
       "epoch": 0.7,
-      "eval_accuracy": 0.6846666666666666,
-      "eval_f1": 0.7591649694501018,
-      "eval_loss": 1.1700917482376099,
-      "eval_precision": 0.6140856672158155,
-      "eval_recall": 0.994,
-      "eval_runtime": 80.6642,
-      "eval_samples_per_second": 37.191,
-      "eval_steps_per_second": 4.649,
+      "eval_accuracy": 0.9123333333333333,
+      "eval_f1": 0.9175290059579806,
+      "eval_loss": 0.6120531558990479,
+      "eval_precision": 0.8661930136175252,
+      "eval_recall": 0.9753333333333334,
+      "eval_runtime": 81.2586,
+      "eval_samples_per_second": 36.919,
+      "eval_steps_per_second": 4.615,
       "step": 1225
     },
     {
       "epoch": 0.72,
       "learning_rate": 1.4000000000000001e-05,
-      "loss": 0.2154,
+      "loss": 0.5469,
       "step": 1260
     },
     {
       "epoch": 0.72,
-      "eval_accuracy": 0.817,
-      "eval_f1": 0.8419234091563489,
-      "eval_loss": 0.48465803265571594,
-      "eval_precision": 0.7410035478966042,
-      "eval_recall": 0.9746666666666667,
-      "eval_runtime": 80.8924,
-      "eval_samples_per_second": 37.086,
-      "eval_steps_per_second": 4.636,
+      "eval_accuracy": 0.9583333333333334,
+      "eval_f1": 0.9584855529724343,
+      "eval_loss": 0.179586261510849,
+      "eval_precision": 0.9549966909331569,
+      "eval_recall": 0.962,
+      "eval_runtime": 81.2049,
+      "eval_samples_per_second": 36.944,
+      "eval_steps_per_second": 4.618,
       "step": 1260
     },
     {
       "epoch": 0.74,
       "learning_rate": 1.3000000000000001e-05,
-      "loss": 0.2415,
+      "loss": 0.1699,
       "step": 1295
     },
     {
       "epoch": 0.74,
-      "eval_accuracy": 0.8613333333333333,
-      "eval_f1": 0.87363304981774,
-      "eval_loss": 0.36877182126045227,
-      "eval_precision": 0.8024553571428571,
-      "eval_recall": 0.9586666666666667,
-      "eval_runtime": 80.8409,
-      "eval_samples_per_second": 37.11,
-      "eval_steps_per_second": 4.639,
+      "eval_accuracy": 0.946,
+      "eval_f1": 0.94375,
+      "eval_loss": 0.24616345763206482,
+      "eval_precision": 0.9847826086956522,
+      "eval_recall": 0.906,
+      "eval_runtime": 83.3002,
+      "eval_samples_per_second": 36.014,
+      "eval_steps_per_second": 4.502,
       "step": 1295
     },
     {
       "epoch": 0.76,
       "learning_rate": 1.2e-05,
-      "loss": 0.2014,
+      "loss": 0.2279,
       "step": 1330
     },
     {
       "epoch": 0.76,
-      "eval_accuracy": 0.7543333333333333,
-      "eval_f1": 0.8008646311807619,
-      "eval_loss": 0.6940145492553711,
-      "eval_precision": 0.6733303044070876,
-      "eval_recall": 0.988,
-      "eval_runtime": 86.2109,
-      "eval_samples_per_second": 34.798,
-      "eval_steps_per_second": 4.35,
+      "eval_accuracy": 0.9576666666666667,
+      "eval_f1": 0.9585644371941273,
+      "eval_loss": 0.19657239317893982,
+      "eval_precision": 0.9386581469648563,
+      "eval_recall": 0.9793333333333333,
+      "eval_runtime": 84.7312,
+      "eval_samples_per_second": 35.406,
+      "eval_steps_per_second": 4.426,
       "step": 1330
     },
     {
       "epoch": 0.78,
       "learning_rate": 1.1000000000000001e-05,
-      "loss": 0.2165,
+      "loss": 0.1383,
       "step": 1365
     },
     {
       "epoch": 0.78,
-      "eval_accuracy": 0.611,
-      "eval_f1": 0.7195385724585437,
-      "eval_loss": 1.6025038957595825,
-      "eval_precision": 0.5625704622322435,
-      "eval_recall": 0.998,
-      "eval_runtime": 81.0854,
-      "eval_samples_per_second": 36.998,
-      "eval_steps_per_second": 4.625,
+      "eval_accuracy": 0.9373333333333334,
+      "eval_f1": 0.9336626676076217,
+      "eval_loss": 0.4285222887992859,
+      "eval_precision": 0.9917541229385307,
+      "eval_recall": 0.882,
+      "eval_runtime": 81.4307,
+      "eval_samples_per_second": 36.841,
+      "eval_steps_per_second": 4.605,
       "step": 1365
     },
     {
       "epoch": 0.8,
       "learning_rate": 1e-05,
-      "loss": 0.2487,
+      "loss": 0.2622,
       "step": 1400
     },
     {
       "epoch": 0.8,
-      "eval_accuracy": 0.73,
-      "eval_f1": 0.7856008470089996,
-      "eval_loss": 0.8998059630393982,
-      "eval_precision": 0.6514486391571555,
-      "eval_recall": 0.9893333333333333,
-      "eval_runtime": 80.4427,
-      "eval_samples_per_second": 37.294,
-      "eval_steps_per_second": 4.662,
+      "eval_accuracy": 0.9606666666666667,
+      "eval_f1": 0.9603227975790181,
+      "eval_loss": 0.22786445915699005,
+      "eval_precision": 0.9687924016282226,
+      "eval_recall": 0.952,
+      "eval_runtime": 81.1895,
+      "eval_samples_per_second": 36.951,
+      "eval_steps_per_second": 4.619,
       "step": 1400
     },
     {
       "epoch": 0.82,
       "learning_rate": 9e-06,
-      "loss": 0.2204,
+      "loss": 0.1573,
       "step": 1435
     },
     {
       "epoch": 0.82,
-      "eval_accuracy": 0.8493333333333334,
-      "eval_f1": 0.8598884066955982,
-      "eval_loss": 0.3596617579460144,
-      "eval_precision": 0.8035921205098494,
-      "eval_recall": 0.9246666666666666,
-      "eval_runtime": 80.5841,
-      "eval_samples_per_second": 37.228,
-      "eval_steps_per_second": 4.654,
+      "eval_accuracy": 0.9636666666666667,
+      "eval_f1": 0.9632873021219266,
+      "eval_loss": 0.1860896348953247,
+      "eval_precision": 0.9734513274336283,
+      "eval_recall": 0.9533333333333334,
+      "eval_runtime": 81.6749,
+      "eval_samples_per_second": 36.731,
+      "eval_steps_per_second": 4.591,
       "step": 1435
     },
     {
       "epoch": 0.84,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 0.1436,
+      "loss": 0.1174,
       "step": 1470
     },
     {
       "epoch": 0.84,
-      "eval_accuracy": 0.5616666666666666,
-      "eval_f1": 0.6951078135868306,
-      "eval_loss": 2.4785938262939453,
-      "eval_precision": 0.5328830430145752,
-      "eval_recall": 0.9993333333333333,
-      "eval_runtime": 80.9085,
-      "eval_samples_per_second": 37.079,
-      "eval_steps_per_second": 4.635,
+      "eval_accuracy": 0.9636666666666667,
+      "eval_f1": 0.9634350888963435,
+      "eval_loss": 0.1978105902671814,
+      "eval_precision": 0.9696151249155975,
+      "eval_recall": 0.9573333333333334,
+      "eval_runtime": 81.6791,
+      "eval_samples_per_second": 36.729,
+      "eval_steps_per_second": 4.591,
       "step": 1470
     },
     {
       "epoch": 0.86,
       "learning_rate": 7.000000000000001e-06,
-      "loss": 0.1431,
+      "loss": 0.1827,
       "step": 1505
     },
     {
       "epoch": 0.86,
-      "eval_accuracy": 0.7606666666666667,
-      "eval_f1": 0.8035030103995622,
-      "eval_loss": 0.7645860314369202,
-      "eval_precision": 0.6815227483751161,
-      "eval_recall": 0.9786666666666667,
-      "eval_runtime": 80.9077,
-      "eval_samples_per_second": 37.079,
-      "eval_steps_per_second": 4.635,
+      "eval_accuracy": 0.9623333333333334,
+      "eval_f1": 0.962219993313273,
+      "eval_loss": 0.215500146150589,
+      "eval_precision": 0.9651240778001341,
+      "eval_recall": 0.9593333333333334,
+      "eval_runtime": 81.1055,
+      "eval_samples_per_second": 36.989,
+      "eval_steps_per_second": 4.624,
       "step": 1505
     },
     {
       "epoch": 0.88,
       "learning_rate": 6e-06,
-      "loss": 0.2083,
+      "loss": 0.2172,
       "step": 1540
     },
     {
       "epoch": 0.88,
-      "eval_accuracy": 0.645,
-      "eval_f1": 0.7372316802368615,
-      "eval_loss": 1.5643061399459839,
-      "eval_precision": 0.5851938895417156,
-      "eval_recall": 0.996,
-      "eval_runtime": 80.4452,
-      "eval_samples_per_second": 37.292,
-      "eval_steps_per_second": 4.662,
+      "eval_accuracy": 0.962,
+      "eval_f1": 0.9615643964935942,
+      "eval_loss": 0.2241593450307846,
+      "eval_precision": 0.9727148703956344,
+      "eval_recall": 0.9506666666666667,
+      "eval_runtime": 81.5602,
+      "eval_samples_per_second": 36.783,
+      "eval_steps_per_second": 4.598,
       "step": 1540
     },
     {
       "epoch": 0.9,
       "learning_rate": 5e-06,
-      "loss": 0.218,
+      "loss": 0.2261,
       "step": 1575
     },
     {
       "epoch": 0.9,
-      "eval_accuracy": 0.6816666666666666,
-      "eval_f1": 0.7570592724497585,
-      "eval_loss": 1.3131792545318604,
-      "eval_precision": 0.6120937885643768,
-      "eval_recall": 0.992,
-      "eval_runtime": 80.8361,
-      "eval_samples_per_second": 37.112,
-      "eval_steps_per_second": 4.639,
+      "eval_accuracy": 0.9643333333333334,
+      "eval_f1": 0.9644636333444038,
+      "eval_loss": 0.19370576739311218,
+      "eval_precision": 0.9609530112508272,
+      "eval_recall": 0.968,
+      "eval_runtime": 81.2706,
+      "eval_samples_per_second": 36.914,
+      "eval_steps_per_second": 4.614,
       "step": 1575
     },
     {
       "epoch": 0.92,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 0.2679,
+      "loss": 0.1769,
       "step": 1610
     },
     {
       "epoch": 0.92,
-      "eval_accuracy": 0.8243333333333334,
-      "eval_f1": 0.845771144278607,
-      "eval_loss": 0.4741792678833008,
-      "eval_precision": 0.7537819509650495,
-      "eval_recall": 0.9633333333333334,
-      "eval_runtime": 80.8168,
-      "eval_samples_per_second": 37.121,
-      "eval_steps_per_second": 4.64,
+      "eval_accuracy": 0.9576666666666667,
+      "eval_f1": 0.9567291311754687,
+      "eval_loss": 0.26763275265693665,
+      "eval_precision": 0.978397212543554,
+      "eval_recall": 0.936,
+      "eval_runtime": 81.1748,
+      "eval_samples_per_second": 36.957,
+      "eval_steps_per_second": 4.62,
       "step": 1610
     },
     {
       "epoch": 0.94,
       "learning_rate": 3e-06,
-      "loss": 0.2218,
+      "loss": 0.4139,
       "step": 1645
     },
     {
       "epoch": 0.94,
-      "eval_accuracy": 0.753,
-      "eval_f1": 0.7996755879967559,
-      "eval_loss": 0.7849438190460205,
-      "eval_precision": 0.6725784447476125,
-      "eval_recall": 0.986,
-      "eval_runtime": 80.4264,
-      "eval_samples_per_second": 37.301,
-      "eval_steps_per_second": 4.663,
+      "eval_accuracy": 0.9673333333333334,
+      "eval_f1": 0.9674634794156707,
+      "eval_loss": 0.1558542400598526,
+      "eval_precision": 0.9636243386243386,
+      "eval_recall": 0.9713333333333334,
+      "eval_runtime": 81.3051,
+      "eval_samples_per_second": 36.898,
+      "eval_steps_per_second": 4.612,
       "step": 1645
     },
     {
       "epoch": 0.96,
       "learning_rate": 2.0000000000000003e-06,
-      "loss": 0.1489,
+      "loss": 0.2209,
       "step": 1680
     },
     {
       "epoch": 0.96,
-      "eval_accuracy": 0.6353333333333333,
-      "eval_f1": 0.7322564855604503,
-      "eval_loss": 1.6629990339279175,
-      "eval_precision": 0.5784996133023975,
-      "eval_recall": 0.9973333333333333,
-      "eval_runtime": 80.5542,
-      "eval_samples_per_second": 37.242,
-      "eval_steps_per_second": 4.655,
+      "eval_accuracy": 0.9706666666666667,
+      "eval_f1": 0.9705488621151273,
+      "eval_loss": 0.1362292617559433,
+      "eval_precision": 0.9744623655913979,
+      "eval_recall": 0.9666666666666667,
+      "eval_runtime": 81.1766,
+      "eval_samples_per_second": 36.956,
+      "eval_steps_per_second": 4.62,
       "step": 1680
     },
     {
       "epoch": 0.98,
       "learning_rate": 1.0000000000000002e-06,
-      "loss": 0.2119,
+      "loss": 0.1792,
       "step": 1715
     },
     {
       "epoch": 0.98,
-      "eval_accuracy": 0.6753333333333333,
-      "eval_f1": 0.7536671724835609,
-      "eval_loss": 1.2936441898345947,
-      "eval_precision": 0.6071719641401793,
-      "eval_recall": 0.9933333333333333,
-      "eval_runtime": 80.4726,
-      "eval_samples_per_second": 37.28,
-      "eval_steps_per_second": 4.66,
+      "eval_accuracy": 0.9696666666666667,
+      "eval_f1": 0.9696363029696363,
+      "eval_loss": 0.13592535257339478,
+      "eval_precision": 0.9706078824315297,
+      "eval_recall": 0.9686666666666667,
+      "eval_runtime": 81.8567,
+      "eval_samples_per_second": 36.649,
+      "eval_steps_per_second": 4.581,
       "step": 1715
     },
     {
       "epoch": 1.0,
       "learning_rate": 0.0,
-      "loss": 0.1341,
+      "loss": 0.1015,
       "step": 1750
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6686666666666666,
-      "eval_f1": 0.7500000000000001,
-      "eval_loss": 1.3422788381576538,
-      "eval_precision": 0.6021809369951535,
-      "eval_recall": 0.994,
-      "eval_runtime": 80.4399,
-      "eval_samples_per_second": 37.295,
-      "eval_steps_per_second": 4.662,
+      "eval_accuracy": 0.9696666666666667,
+      "eval_f1": 0.9696160267111853,
+      "eval_loss": 0.1375236213207245,
+      "eval_precision": 0.9712374581939799,
+      "eval_recall": 0.968,
+      "eval_runtime": 81.3216,
+      "eval_samples_per_second": 36.891,
+      "eval_steps_per_second": 4.611,
       "step": 1750
     }
   ],