End of training

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 10.0,
     "eval_accuracy": 0.9306930899620056,
-    "eval_f1": 0.9797514033680834,
-    "eval_loss": 0.1402171403169632,
-    "eval_runtime": 0.4188,
+    "eval_f1": 0.8582166199813259,
+    "eval_loss": 0.3003327250480652,
+    "eval_runtime": 0.471,
     "eval_samples": 101,
-    "eval_samples_per_second": 241.172,
-    "eval_steps_per_second": 31.042,
-    "train_loss": 0.13698372087817182,
-    "train_runtime": 202.1369,
+    "eval_samples_per_second": 214.442,
+    "eval_steps_per_second": 27.602,
+    "train_loss": 0.13365532121531207,
+    "train_runtime": 203.1053,
     "train_samples": 908,
-    "train_samples_per_second": 44.92,
-    "train_steps_per_second": 2.82
+    "train_samples_per_second": 44.706,
+    "train_steps_per_second": 2.806
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_f1": 0.9797514033680834,
-    "eval_loss": 0.1402171403169632,
-    "eval_runtime": 0.4188,
+    "eval_f1": 0.8582166199813259,
+    "eval_loss": 0.3003327250480652,
+    "eval_runtime": 0.471,
     "eval_samples": 101,
-    "eval_samples_per_second": 241.172,
-    "eval_steps_per_second": 31.042
+    "eval_samples_per_second": 214.442,
+    "eval_steps_per_second": 27.602
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d8f6962966f9657e2a209ddb25ec343a7ee70303a14f7b8f5b08ce7b72bc3c0
+oid sha256:5616bb838fcbc2cd5c74925cf7f528cdb45e4a737239a518c9be6426f937b6f0
 size 713322669

runs/1636580941.1874316/events.out.tfevents.1636580941.5690110ef819.20333.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f7c0badaacee2d44801c0445a045a83a851dd9148199622f37f0a838c0e682e
+size 4480

runs/events.out.tfevents.1636580941.5690110ef819.20333.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9fa34f0817f7ee033dc2fb8a127001bad547178b0e620a28d70c3bab2f75406
+size 25189

runs/events.out.tfevents.1636581197.5690110ef819.20333.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfce0d876b55a0c01ab5e62cbd8bba3ca6141d1715e48af30030c45c0be15144
+size 357

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 0.13698372087817182,
-    "train_runtime": 202.1369,
+    "train_loss": 0.13365532121531207,
+    "train_runtime": 203.1053,
     "train_samples": 908,
-    "train_samples_per_second": 44.92,
-    "train_steps_per_second": 2.82
+    "train_samples_per_second": 44.706,
+    "train_steps_per_second": 2.806
 }

trainer_state.json

@@ -10,499 +10,499 @@
     {
       "epoch": 0.02,
       "learning_rate": 1.4285714285714286e-06,
-      "loss": 1.1535,
+      "loss": 1.0431,
       "step": 1
     },
     {
       "epoch": 0.09,
       "learning_rate": 7.142857142857143e-06,
-      "loss": 1.0482,
+      "loss": 1.0528,
       "step": 5
     },
     {
       "epoch": 0.18,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.9675,
+      "loss": 0.9619,
       "step": 10
     },
     {
       "epoch": 0.26,
       "learning_rate": 2.1428571428571428e-05,
-      "loss": 0.8126,
+      "loss": 0.845,
       "step": 15
     },
     {
       "epoch": 0.35,
       "learning_rate": 2.857142857142857e-05,
-      "loss": 0.7321,
+      "loss": 0.7895,
       "step": 20
     },
     {
       "epoch": 0.44,
       "learning_rate": 3.571428571428572e-05,
-      "loss": 0.7198,
+      "loss": 0.6763,
       "step": 25
     },
     {
       "epoch": 0.53,
       "learning_rate": 4.2857142857142856e-05,
-      "loss": 0.785,
+      "loss": 0.6805,
       "step": 30
     },
     {
       "epoch": 0.61,
       "learning_rate": 5e-05,
-      "loss": 0.7282,
+      "loss": 0.658,
       "step": 35
     },
     {
       "epoch": 0.7,
       "learning_rate": 4.998922515567496e-05,
-      "loss": 0.7238,
+      "loss": 0.7479,
       "step": 40
     },
     {
       "epoch": 0.79,
       "learning_rate": 4.995690991048146e-05,
-      "loss": 0.6667,
+      "loss": 0.722,
       "step": 45
     },
     {
       "epoch": 0.88,
       "learning_rate": 4.99030821197584e-05,
-      "loss": 0.6404,
+      "loss": 0.6369,
       "step": 50
     },
     {
       "epoch": 0.96,
       "learning_rate": 4.982778818239101e-05,
-      "loss": 0.4274,
+      "loss": 0.4374,
       "step": 55
     },
     {
       "epoch": 1.0,
-      "eval_f1": 0.7878385848535102,
-      "eval_loss": 0.4914661943912506,
-      "eval_runtime": 0.6719,
-      "eval_samples_per_second": 150.315,
-      "eval_steps_per_second": 19.348,
+      "eval_f1": 0.5522201375859912,
+      "eval_loss": 0.42813247442245483,
+      "eval_runtime": 0.7559,
+      "eval_samples_per_second": 133.614,
+      "eval_steps_per_second": 17.198,
       "step": 57
     },
     {
       "epoch": 1.05,
       "learning_rate": 4.97310930008156e-05,
-      "loss": 0.5671,
+      "loss": 0.5211,
       "step": 60
     },
     {
       "epoch": 1.14,
       "learning_rate": 4.961307992507443e-05,
-      "loss": 0.3926,
+      "loss": 0.4878,
       "step": 65
     },
     {
       "epoch": 1.23,
       "learning_rate": 4.947385068096907e-05,
-      "loss": 0.2036,
+      "loss": 0.2413,
       "step": 70
     },
     {
       "epoch": 1.32,
       "learning_rate": 4.9313525282373974e-05,
-      "loss": 0.3489,
+      "loss": 0.371,
       "step": 75
     },
     {
       "epoch": 1.4,
       "learning_rate": 4.9132241927786035e-05,
-      "loss": 0.3307,
+      "loss": 0.388,
       "step": 80
     },
     {
       "epoch": 1.49,
       "learning_rate": 4.893015688119921e-05,
-      "loss": 0.2562,
+      "loss": 0.3017,
       "step": 85
     },
     {
       "epoch": 1.58,
       "learning_rate": 4.870744433740688e-05,
-      "loss": 0.322,
+      "loss": 0.4045,
       "step": 90
     },
     {
       "epoch": 1.67,
       "learning_rate": 4.8464296271848155e-05,
-      "loss": 0.2665,
+      "loss": 0.1947,
       "step": 95
     },
     {
       "epoch": 1.75,
       "learning_rate": 4.8200922275127355e-05,
-      "loss": 0.249,
+      "loss": 0.1576,
       "step": 100
     },
     {
       "epoch": 1.84,
       "learning_rate": 4.7917549372349616e-05,
-      "loss": 0.2196,
+      "loss": 0.1537,
       "step": 105
     },
     {
       "epoch": 1.93,
       "learning_rate": 4.761442182742799e-05,
-      "loss": 0.2177,
+      "loss": 0.2038,
       "step": 110
     },
     {
       "epoch": 2.0,
-      "eval_f1": 0.848931623931624,
-      "eval_loss": 0.4218161702156067,
-      "eval_runtime": 0.4386,
-      "eval_samples_per_second": 230.292,
-      "eval_steps_per_second": 29.641,
+      "eval_f1": 0.6002116402116403,
+      "eval_loss": 0.33254939317703247,
+      "eval_runtime": 0.4237,
+      "eval_samples_per_second": 238.366,
+      "eval_steps_per_second": 30.681,
       "step": 114
     },
     {
       "epoch": 2.02,
       "learning_rate": 4.7291800932531064e-05,
-      "loss": 0.2602,
+      "loss": 0.0917,
       "step": 115
     },
     {
       "epoch": 2.11,
       "learning_rate": 4.694996478285231e-05,
-      "loss": 0.5002,
+      "loss": 0.1923,
       "step": 120
     },
     {
       "epoch": 2.19,
       "learning_rate": 4.6589208036895535e-05,
-      "loss": 0.2037,
+      "loss": 0.0289,
       "step": 125
     },
     {
       "epoch": 2.28,
       "learning_rate": 4.620984166248288e-05,
-      "loss": 0.1931,
+      "loss": 0.4299,
       "step": 130
     },
     {
       "epoch": 2.37,
       "learning_rate": 4.581219266870446e-05,
-      "loss": 0.1799,
+      "loss": 0.1862,
       "step": 135
     },
     {
       "epoch": 2.46,
       "learning_rate": 4.53966038240406e-05,
-      "loss": 0.1665,
+      "loss": 0.1792,
       "step": 140
     },
     {
       "epoch": 2.54,
       "learning_rate": 4.496343336089965e-05,
-      "loss": 0.1888,
+      "loss": 0.198,
       "step": 145
     },
     {
       "epoch": 2.63,
       "learning_rate": 4.4513054666826146e-05,
-      "loss": 0.0738,
+      "loss": 0.2583,
       "step": 150
     },
     {
       "epoch": 2.72,
       "learning_rate": 4.4045855962645363e-05,
-      "loss": 0.1971,
+      "loss": 0.0596,
       "step": 155
     },
     {
       "epoch": 2.81,
       "learning_rate": 4.3562239967821805e-05,
-      "loss": 0.2111,
+      "loss": 0.1632,
       "step": 160
     },
     {
       "epoch": 2.89,
       "learning_rate": 4.306262355332006e-05,
-      "loss": 0.0525,
+      "loss": 0.0914,
       "step": 165
     },
     {
       "epoch": 2.98,
       "learning_rate": 4.254743738226721e-05,
-      "loss": 0.1154,
+      "loss": 0.125,
       "step": 170
     },
     {
       "epoch": 3.0,
-      "eval_f1": 0.959920634920635,
-      "eval_loss": 0.17385226488113403,
-      "eval_runtime": 1.7225,
-      "eval_samples_per_second": 58.637,
-      "eval_steps_per_second": 7.547,
+      "eval_f1": 0.6362258953168044,
+      "eval_loss": 0.19313393533229828,
+      "eval_runtime": 1.7454,
+      "eval_samples_per_second": 57.867,
+      "eval_steps_per_second": 7.448,
       "step": 171
     },
     {
       "epoch": 3.07,
       "learning_rate": 4.201712553872658e-05,
-      "loss": 0.2055,
+      "loss": 0.0745,
       "step": 175
     },
     {
       "epoch": 3.16,
       "learning_rate": 4.147214514490278e-05,
-      "loss": 0.0767,
+      "loss": 0.1091,
       "step": 180
     },
     {
       "epoch": 3.25,
       "learning_rate": 4.0912965967108125e-05,
-      "loss": 0.0224,
+      "loss": 0.0114,
       "step": 185
     },
     {
       "epoch": 3.33,
       "learning_rate": 4.034007001082985e-05,
-      "loss": 0.1034,
+      "loss": 0.0886,
       "step": 190
     },
     {
       "epoch": 3.42,
       "learning_rate": 3.975395110524742e-05,
-      "loss": 0.0192,
+      "loss": 0.08,
       "step": 195
     },
     {
       "epoch": 3.51,
       "learning_rate": 3.9155114477557933e-05,
-      "loss": 0.069,
+      "loss": 0.1271,
       "step": 200
     },
     {
       "epoch": 3.6,
       "learning_rate": 3.854407631747654e-05,
-      "loss": 0.0203,
+      "loss": 0.0584,
       "step": 205
     },
     {
       "epoch": 3.68,
       "learning_rate": 3.792136333228735e-05,
-      "loss": 0.2089,
+      "loss": 0.1567,
       "step": 210
     },
     {
       "epoch": 3.77,
       "learning_rate": 3.728751229282836e-05,
-      "loss": 0.1181,
+      "loss": 0.0222,
       "step": 215
     },
     {
       "epoch": 3.86,
       "learning_rate": 3.664306957080159e-05,
-      "loss": 0.0077,
+      "loss": 0.2284,
       "step": 220
     },
     {
       "epoch": 3.95,
       "learning_rate": 3.598859066780754e-05,
-      "loss": 0.1192,
+      "loss": 0.0135,
       "step": 225
     },
     {
       "epoch": 4.0,
-      "eval_f1": 0.9592413236481033,
-      "eval_loss": 0.15608876943588257,
-      "eval_runtime": 0.4317,
-      "eval_samples_per_second": 233.965,
-      "eval_steps_per_second": 30.114,
+      "eval_f1": 0.7928141817030706,
+      "eval_loss": 0.21502041816711426,
+      "eval_runtime": 0.4545,
+      "eval_samples_per_second": 222.231,
+      "eval_steps_per_second": 28.604,
       "step": 228
     },
     {
       "epoch": 4.04,
       "learning_rate": 3.5324639736509714e-05,
-      "loss": 0.0058,
+      "loss": 0.0824,
       "step": 230
     },
     {
       "epoch": 4.12,
       "learning_rate": 3.4651789094342044e-05,
-      "loss": 0.038,
+      "loss": 0.0428,
       "step": 235
     },
     {
       "epoch": 4.21,
       "learning_rate": 3.39706187301784e-05,
-      "loss": 0.0024,
+      "loss": 0.0037,
       "step": 240
     },
     {
       "epoch": 4.3,
       "learning_rate": 3.3281715804389403e-05,
-      "loss": 0.0277,
+      "loss": 0.0033,
       "step": 245
     },
     {
       "epoch": 4.39,
       "learning_rate": 3.258567414271748e-05,
-      "loss": 0.0023,
+      "loss": 0.003,
       "step": 250
     },
     {
       "epoch": 4.47,
       "learning_rate": 3.18830937244065e-05,
-      "loss": 0.145,
+      "loss": 0.0022,
       "step": 255
     },
     {
       "epoch": 4.56,
       "learning_rate": 3.117458016502711e-05,
-      "loss": 0.0818,
+      "loss": 0.0018,
       "step": 260
     },
     {
       "epoch": 4.65,
       "learning_rate": 3.046074419444366e-05,
-      "loss": 0.0089,
+      "loss": 0.0015,
       "step": 265
     },
     {
       "epoch": 4.74,
       "learning_rate": 2.9742201130372693e-05,
-      "loss": 0.0026,
+      "loss": 0.0013,
       "step": 270
     },
     {
       "epoch": 4.82,
       "learning_rate": 2.901957034798671e-05,
-      "loss": 0.0032,
+      "loss": 0.1622,
       "step": 275
     },
     {
       "epoch": 4.91,
       "learning_rate": 2.8293474746020472e-05,
-      "loss": 0.0339,
+      "loss": 0.0093,
       "step": 280
     },
     {
       "epoch": 5.0,
       "learning_rate": 2.756454020984009e-05,
-      "loss": 0.0022,
+      "loss": 0.0939,
       "step": 285
     },
     {
       "epoch": 5.0,
-      "eval_f1": 0.9592413236481033,
-      "eval_loss": 0.09981890022754669,
-      "eval_runtime": 0.4318,
-      "eval_samples_per_second": 233.88,
-      "eval_steps_per_second": 30.103,
+      "eval_f1": 0.7998317914213624,
+      "eval_loss": 0.14565160870552063,
+      "eval_runtime": 0.4475,
+      "eval_samples_per_second": 225.691,
+      "eval_steps_per_second": 29.049,
       "step": 285
     },
     {
       "epoch": 5.09,
       "learning_rate": 2.68333950719376e-05,
-      "loss": 0.0023,
+      "loss": 0.0806,
       "step": 290
     },
     {
       "epoch": 5.18,
       "learning_rate": 2.6100669570316195e-05,
-      "loss": 0.0022,
+      "loss": 0.0034,
       "step": 295
     },
     {
       "epoch": 5.26,
       "learning_rate": 2.5366995305232916e-05,
-      "loss": 0.0016,
+      "loss": 0.004,
       "step": 300
     },
     {
       "epoch": 5.35,
       "learning_rate": 2.463300469476709e-05,
-      "loss": 0.0012,
+      "loss": 0.037,
       "step": 305
     },
     {
       "epoch": 5.44,
       "learning_rate": 2.3899330429683807e-05,
-      "loss": 0.0011,
+      "loss": 0.0021,
       "step": 310
     },
     {
       "epoch": 5.53,
       "learning_rate": 2.3166604928062406e-05,
-      "loss": 0.002,
+      "loss": 0.0016,
       "step": 315
     },
     {
       "epoch": 5.61,
       "learning_rate": 2.243545979015992e-05,
-      "loss": 0.0051,
+      "loss": 0.0016,
       "step": 320
     },
     {
       "epoch": 5.7,
       "learning_rate": 2.1706525253979534e-05,
-      "loss": 0.0169,
+      "loss": 0.0014,
       "step": 325
     },
     {
       "epoch": 5.79,
       "learning_rate": 2.0980429652013297e-05,
-      "loss": 0.0925,
+      "loss": 0.0013,
       "step": 330
     },
     {
       "epoch": 5.88,
       "learning_rate": 2.025779886962731e-05,
-      "loss": 0.0008,
+      "loss": 0.0011,
       "step": 335
     },
     {
       "epoch": 5.96,
       "learning_rate": 1.9539255805556344e-05,
-      "loss": 0.0009,
+      "loss": 0.001,
       "step": 340
     },
     {
       "epoch": 6.0,
-      "eval_f1": 0.9797514033680834,
-      "eval_loss": 0.10345583409070969,
-      "eval_runtime": 0.411,
-      "eval_samples_per_second": 245.716,
-      "eval_steps_per_second": 31.627,
+      "eval_f1": 0.8066695032443358,
+      "eval_loss": 0.2502600848674774,
+      "eval_runtime": 0.4423,
+      "eval_samples_per_second": 228.329,
+      "eval_steps_per_second": 29.389,
       "step": 342
     },
     {
       "epoch": 6.05,
       "learning_rate": 1.8825419834972902e-05,
-      "loss": 0.0508,
+      "loss": 0.0011,
       "step": 345
     },
     {
       "epoch": 6.14,
       "learning_rate": 1.811690627559351e-05,
-      "loss": 0.0014,
+      "loss": 0.0011,
       "step": 350
     },
     {
       "epoch": 6.23,
       "learning_rate": 1.7414325857282526e-05,
-      "loss": 0.0014,
+      "loss": 0.0009,
       "step": 355
     },
     {
       "epoch": 6.32,
       "learning_rate": 1.6718284195610606e-05,
-      "loss": 0.0014,
+      "loss": 0.0009,
       "step": 360
     },
     {
       "epoch": 6.4,
       "learning_rate": 1.6029381269821604e-05,
-      "loss": 0.0011,
+      "loss": 0.0007,
       "step": 365
     },
     {
@@ -514,46 +514,46 @@
     {
       "epoch": 6.58,
       "learning_rate": 1.4675360263490295e-05,
-      "loss": 0.0009,
+      "loss": 0.0605,
       "step": 375
     },
     {
       "epoch": 6.67,
       "learning_rate": 1.4011409332192472e-05,
-      "loss": 0.0043,
+      "loss": 0.0008,
       "step": 380
     },
     {
       "epoch": 6.75,
       "learning_rate": 1.335693042919841e-05,
-      "loss": 0.0006,
+      "loss": 0.0024,
       "step": 385
     },
     {
       "epoch": 6.84,
       "learning_rate": 1.2712487707171645e-05,
-      "loss": 0.0246,
+      "loss": 0.0009,
       "step": 390
     },
     {
       "epoch": 6.93,
       "learning_rate": 1.2078636667712649e-05,
-      "loss": 0.0006,
+      "loss": 0.0008,
       "step": 395
     },
     {
       "epoch": 7.0,
-      "eval_f1": 0.9797514033680834,
-      "eval_loss": 0.12670956552028656,
-      "eval_runtime": 0.4031,
-      "eval_samples_per_second": 250.582,
-      "eval_steps_per_second": 32.253,
+      "eval_f1": 0.8582166199813259,
+      "eval_loss": 0.309122771024704,
+      "eval_runtime": 0.4346,
+      "eval_samples_per_second": 232.408,
+      "eval_steps_per_second": 29.914,
       "step": 399
     },
     {
       "epoch": 7.02,
       "learning_rate": 1.1455923682523475e-05,
-      "loss": 0.0007,
+      "loss": 0.0011,
       "step": 400
     },
     {
@@ -571,13 +571,13 @@
     {
       "epoch": 7.28,
       "learning_rate": 9.659929989170154e-06,
-      "loss": 0.0006,
+      "loss": 0.0007,
       "step": 415
     },
     {
       "epoch": 7.37,
       "learning_rate": 9.087034032891883e-06,
-      "loss": 0.0007,
+      "loss": 0.0005,
       "step": 420
     },
     {
@@ -589,13 +589,13 @@
     {
       "epoch": 7.54,
       "learning_rate": 7.982874461273438e-06,
-      "loss": 0.0005,
+      "loss": 0.0006,
       "step": 430
     },
     {
       "epoch": 7.63,
       "learning_rate": 7.452562617732794e-06,
-      "loss": 0.0625,
+      "loss": 0.0007,
       "step": 435
     },
     {
@@ -613,7 +613,7 @@
     {
       "epoch": 7.89,
       "learning_rate": 5.954144037354645e-06,
-      "loss": 0.0006,
+      "loss": 0.0005,
       "step": 450
     },
     {
@@ -624,17 +624,17 @@
     },
     {
       "epoch": 8.0,
-      "eval_f1": 0.9797514033680834,
-      "eval_loss": 0.1428430825471878,
-      "eval_runtime": 0.4454,
-      "eval_samples_per_second": 226.778,
-      "eval_steps_per_second": 29.189,
+      "eval_f1": 0.8582166199813259,
+      "eval_loss": 0.287475049495697,
+      "eval_runtime": 0.4503,
+      "eval_samples_per_second": 224.273,
+      "eval_steps_per_second": 28.867,
       "step": 456
     },
     {
       "epoch": 8.07,
       "learning_rate": 5.036566639100351e-06,
-      "loss": 0.0096,
+      "loss": 0.0006,
       "step": 460
     },
     {
@@ -646,13 +646,13 @@
     {
       "epoch": 8.25,
       "learning_rate": 4.187807331295549e-06,
-      "loss": 0.0005,
+      "loss": 0.0004,
       "step": 470
     },
     {
       "epoch": 8.33,
       "learning_rate": 3.7901583375171273e-06,
-      "loss": 0.0005,
+      "loss": 0.0004,
       "step": 475
     },
     {
@@ -682,7 +682,7 @@
     {
       "epoch": 8.77,
       "learning_rate": 2.0824506276503897e-06,
-      "loss": 0.0005,
+      "loss": 0.0004,
       "step": 500
     },
     {
@@ -694,34 +694,34 @@
     {
       "epoch": 8.95,
       "learning_rate": 1.5357037281518522e-06,
-      "loss": 0.0005,
+      "loss": 0.0004,
       "step": 510
     },
     {
       "epoch": 9.0,
-      "eval_f1": 0.9797514033680834,
-      "eval_loss": 0.13912993669509888,
-      "eval_runtime": 0.43,
-      "eval_samples_per_second": 234.9,
-      "eval_steps_per_second": 30.235,
+      "eval_f1": 0.8582166199813259,
+      "eval_loss": 0.29764121770858765,
+      "eval_runtime": 0.4478,
+      "eval_samples_per_second": 225.548,
+      "eval_steps_per_second": 29.031,
       "step": 513
     },
     {
       "epoch": 9.04,
       "learning_rate": 1.2925556625931173e-06,
-      "loss": 0.0004,
+      "loss": 0.0005,
       "step": 515
     },
     {
       "epoch": 9.12,
       "learning_rate": 1.0698431188007952e-06,
-      "loss": 0.0005,
+      "loss": 0.0004,
       "step": 520
     },
     {
       "epoch": 9.21,
       "learning_rate": 8.677580722139672e-07,
-      "loss": 0.0004,
+      "loss": 0.0005,
       "step": 525
     },
     {
@@ -733,13 +733,13 @@
     {
       "epoch": 9.39,
       "learning_rate": 5.261493190309303e-07,
-      "loss": 0.0005,
+      "loss": 0.0004,
       "step": 535
     },
     {
       "epoch": 9.47,
       "learning_rate": 3.8692007492557024e-07,
-      "loss": 0.0004,
+      "loss": 0.0005,
       "step": 540
     },
     {
@@ -775,31 +775,31 @@
     {
       "epoch": 10.0,
       "learning_rate": 0.0,
-      "loss": 0.0005,
+      "loss": 0.0007,
       "step": 570
     },
     {
       "epoch": 10.0,
-      "eval_f1": 0.9797514033680834,
-      "eval_loss": 0.1402171403169632,
-      "eval_runtime": 1.7273,
-      "eval_samples_per_second": 58.472,
-      "eval_steps_per_second": 7.526,
+      "eval_f1": 0.8582166199813259,
+      "eval_loss": 0.3003327250480652,
+      "eval_runtime": 1.7359,
+      "eval_samples_per_second": 58.182,
+      "eval_steps_per_second": 7.489,
       "step": 570
     },
     {
       "epoch": 10.0,
       "step": 570,
-      "total_flos": 200268366788952.0,
-      "train_loss": 0.13698372087817182,
-      "train_runtime": 202.1369,
-      "train_samples_per_second": 44.92,
-      "train_steps_per_second": 2.82
+      "total_flos": 196303165360848.0,
+      "train_loss": 0.13365532121531207,
+      "train_runtime": 203.1053,
+      "train_samples_per_second": 44.706,
+      "train_steps_per_second": 2.806
     }
   ],
   "max_steps": 570,
   "num_train_epochs": 10,
-  "total_flos": 200268366788952.0,
+  "total_flos": 196303165360848.0,
   "trial_name": null,
   "trial_params": null
 }