End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 9.0,
-    "eval_accuracy": 0.8969686033922771,
-    "eval_loss": 0.3010924756526947,
-    "eval_runtime": 59.407,
-    "eval_samples_per_second": 93.289,
-    "eval_steps_per_second": 2.929,
-    "total_flos": 9.017904477803397e+18,
-    "train_loss": 0.6144383929350196,
-    "train_runtime": 3112.0693,
-    "train_samples_per_second": 415.479,
-    "train_steps_per_second": 3.245
+    "epoch": 11.0,
+    "eval_accuracy": 0.9157343919162757,
+    "eval_loss": 0.25571852922439575,
+    "eval_runtime": 59.9611,
+    "eval_samples_per_second": 92.427,
+    "eval_steps_per_second": 2.902,
+    "total_flos": 1.1021883250648596e+19,
+    "train_loss": 0.1948466427711406,
+    "train_runtime": 3819.4144,
+    "train_samples_per_second": 338.534,
+    "train_steps_per_second": 2.644
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.0,
-    "eval_accuracy": 0.8969686033922771,
-    "eval_loss": 0.3010924756526947,
-    "eval_runtime": 59.407,
-    "eval_samples_per_second": 93.289,
-    "eval_steps_per_second": 2.929
+    "epoch": 11.0,
+    "eval_accuracy": 0.9157343919162757,
+    "eval_loss": 0.25571852922439575,
+    "eval_runtime": 59.9611,
+    "eval_samples_per_second": 92.427,
+    "eval_steps_per_second": 2.902
 }

runs/Apr03_13-59-46_49255a4cc0b5/events.out.tfevents.1649002554.49255a4cc0b5.97.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e68184052905c93994725f07ce3f3a90f962d305ae3a3bc4fabffa2e6dfef96
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.0,
-    "total_flos": 9.017904477803397e+18,
-    "train_loss": 0.6144383929350196,
-    "train_runtime": 3112.0693,
-    "train_samples_per_second": 415.479,
-    "train_steps_per_second": 3.245
+    "epoch": 11.0,
+    "total_flos": 1.1021883250648596e+19,
+    "train_loss": 0.1948466427711406,
+    "train_runtime": 3819.4144,
+    "train_samples_per_second": 338.534,
+    "train_steps_per_second": 2.644
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.8969686033922771,
+  "best_metric": 0.9157343919162757,
   "best_model_checkpoint": "violation-classification-bantai-vit-v100ep/checkpoint-808",
-  "epoch": 8.99753086419753,
-  "global_step": 909,
+  "epoch": 10.99753086419753,
+  "global_step": 1111,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -10,637 +10,781 @@
     {
       "epoch": 0.1,
       "learning_rate": 4.950495049504951e-07,
-      "loss": 1.3728,
+      "loss": 0.2829,
       "step": 10
     },
     {
       "epoch": 0.2,
       "learning_rate": 9.900990099009902e-07,
-      "loss": 1.37,
+      "loss": 0.2759,
       "step": 20
     },
     {
       "epoch": 0.3,
       "learning_rate": 1.4851485148514852e-06,
-      "loss": 1.3698,
+      "loss": 0.2657,
       "step": 30
     },
     {
       "epoch": 0.4,
       "learning_rate": 1.9801980198019803e-06,
-      "loss": 1.3582,
+      "loss": 0.2642,
       "step": 40
     },
     {
       "epoch": 0.49,
       "learning_rate": 2.4752475247524753e-06,
-      "loss": 1.348,
+      "loss": 0.2586,
       "step": 50
     },
     {
       "epoch": 0.59,
       "learning_rate": 2.9702970297029703e-06,
-      "loss": 1.3355,
+      "loss": 0.2344,
       "step": 60
     },
     {
       "epoch": 0.69,
       "learning_rate": 3.4653465346534657e-06,
-      "loss": 1.3186,
+      "loss": 0.2868,
       "step": 70
     },
     {
       "epoch": 0.79,
       "learning_rate": 3.960396039603961e-06,
-      "loss": 1.3057,
+      "loss": 0.2831,
       "step": 80
     },
     {
       "epoch": 0.89,
       "learning_rate": 4.455445544554456e-06,
-      "loss": 1.2756,
+      "loss": 0.2574,
       "step": 90
     },
     {
       "epoch": 0.99,
       "learning_rate": 4.950495049504951e-06,
-      "loss": 1.2596,
+      "loss": 0.2811,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5615301335258029,
-      "eval_loss": 1.222981333732605,
-      "eval_runtime": 59.3749,
-      "eval_samples_per_second": 93.339,
-      "eval_steps_per_second": 2.931,
+      "eval_accuracy": 0.9027426921688921,
+      "eval_loss": 0.28554767370224,
+      "eval_runtime": 59.413,
+      "eval_samples_per_second": 93.279,
+      "eval_steps_per_second": 2.929,
       "step": 101
     },
     {
       "epoch": 1.09,
       "learning_rate": 5.445544554455446e-06,
-      "loss": 1.2563,
+      "loss": 0.2769,
       "step": 110
     },
     {
       "epoch": 1.19,
       "learning_rate": 5.940594059405941e-06,
-      "loss": 1.1843,
+      "loss": 0.2356,
       "step": 120
     },
     {
       "epoch": 1.29,
       "learning_rate": 6.4356435643564364e-06,
-      "loss": 1.1572,
+      "loss": 0.2493,
       "step": 130
     },
     {
       "epoch": 1.39,
       "learning_rate": 6.9306930693069314e-06,
-      "loss": 1.1153,
+      "loss": 0.2455,
       "step": 140
     },
     {
       "epoch": 1.48,
       "learning_rate": 7.4257425742574256e-06,
-      "loss": 1.0653,
+      "loss": 0.2305,
       "step": 150
     },
     {
       "epoch": 1.58,
       "learning_rate": 7.920792079207921e-06,
-      "loss": 1.0119,
+      "loss": 0.2439,
       "step": 160
     },
     {
       "epoch": 1.68,
       "learning_rate": 8.415841584158417e-06,
-      "loss": 0.9734,
+      "loss": 0.2493,
       "step": 170
     },
     {
       "epoch": 1.78,
       "learning_rate": 8.910891089108911e-06,
-      "loss": 0.9144,
+      "loss": 0.2357,
       "step": 180
     },
     {
       "epoch": 1.88,
       "learning_rate": 9.405940594059407e-06,
-      "loss": 0.9037,
+      "loss": 0.2709,
       "step": 190
     },
     {
       "epoch": 1.98,
       "learning_rate": 9.900990099009901e-06,
-      "loss": 0.8527,
+      "loss": 0.2382,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6840490797546013,
-      "eval_loss": 0.8233989477157593,
-      "eval_runtime": 59.669,
-      "eval_samples_per_second": 92.879,
-      "eval_steps_per_second": 2.916,
+      "eval_accuracy": 0.908516780945507,
+      "eval_loss": 0.27634507417678833,
+      "eval_runtime": 58.9068,
+      "eval_samples_per_second": 94.081,
+      "eval_steps_per_second": 2.954,
       "step": 202
     },
     {
       "epoch": 2.08,
       "learning_rate": 1.0396039603960395e-05,
-      "loss": 0.8402,
+      "loss": 0.2447,
       "step": 210
     },
     {
       "epoch": 2.18,
       "learning_rate": 1.0891089108910891e-05,
-      "loss": 0.7847,
+      "loss": 0.2369,
       "step": 220
     },
     {
       "epoch": 2.28,
       "learning_rate": 1.1386138613861387e-05,
-      "loss": 0.7738,
+      "loss": 0.2213,
       "step": 230
     },
     {
       "epoch": 2.38,
       "learning_rate": 1.1881188118811881e-05,
-      "loss": 0.7346,
+      "loss": 0.2433,
       "step": 240
     },
     {
       "epoch": 2.47,
       "learning_rate": 1.2376237623762377e-05,
-      "loss": 0.6992,
+      "loss": 0.2184,
       "step": 250
     },
     {
       "epoch": 2.57,
       "learning_rate": 1.2871287128712873e-05,
-      "loss": 0.695,
+      "loss": 0.22,
       "step": 260
     },
     {
       "epoch": 2.67,
       "learning_rate": 1.3366336633663367e-05,
-      "loss": 0.6715,
+      "loss": 0.2111,
       "step": 270
     },
     {
       "epoch": 2.77,
       "learning_rate": 1.3861386138613863e-05,
-      "loss": 0.6683,
+      "loss": 0.2134,
       "step": 280
     },
     {
       "epoch": 2.87,
       "learning_rate": 1.4356435643564355e-05,
-      "loss": 0.6617,
+      "loss": 0.222,
       "step": 290
     },
     {
       "epoch": 2.97,
       "learning_rate": 1.4851485148514851e-05,
-      "loss": 0.6375,
+      "loss": 0.2361,
       "step": 300
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.784554312522555,
-      "eval_loss": 0.6000781655311584,
-      "eval_runtime": 59.2996,
-      "eval_samples_per_second": 93.458,
-      "eval_steps_per_second": 2.934,
+      "eval_accuracy": 0.9108625045110068,
+      "eval_loss": 0.2605169415473938,
+      "eval_runtime": 59.2161,
+      "eval_samples_per_second": 93.589,
+      "eval_steps_per_second": 2.938,
       "step": 303
     },
     {
       "epoch": 3.07,
       "learning_rate": 1.534653465346535e-05,
-      "loss": 0.6349,
+      "loss": 0.2091,
       "step": 310
     },
     {
       "epoch": 3.17,
       "learning_rate": 1.5841584158415843e-05,
-      "loss": 0.5924,
+      "loss": 0.2363,
       "step": 320
     },
     {
       "epoch": 3.27,
       "learning_rate": 1.6336633663366337e-05,
-      "loss": 0.5905,
+      "loss": 0.2161,
       "step": 330
     },
     {
       "epoch": 3.37,
       "learning_rate": 1.6831683168316834e-05,
-      "loss": 0.5591,
+      "loss": 0.2232,
       "step": 340
     },
     {
       "epoch": 3.46,
       "learning_rate": 1.7326732673267325e-05,
-      "loss": 0.582,
+      "loss": 0.2071,
       "step": 350
     },
     {
       "epoch": 3.56,
       "learning_rate": 1.7821782178217823e-05,
-      "loss": 0.553,
+      "loss": 0.1942,
       "step": 360
     },
     {
       "epoch": 3.66,
       "learning_rate": 1.8316831683168317e-05,
-      "loss": 0.5489,
+      "loss": 0.2114,
       "step": 370
     },
     {
       "epoch": 3.76,
       "learning_rate": 1.8811881188118814e-05,
-      "loss": 0.5711,
+      "loss": 0.2158,
       "step": 380
     },
     {
       "epoch": 3.86,
       "learning_rate": 1.930693069306931e-05,
-      "loss": 0.5161,
+      "loss": 0.1908,
       "step": 390
     },
     {
       "epoch": 3.96,
       "learning_rate": 1.9801980198019803e-05,
-      "loss": 0.555,
+      "loss": 0.196,
       "step": 400
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.817755322988091,
-      "eval_loss": 0.5038326978683472,
-      "eval_runtime": 59.2865,
-      "eval_samples_per_second": 93.478,
-      "eval_steps_per_second": 2.935,
+      "eval_accuracy": 0.911042944785276,
+      "eval_loss": 0.2652387320995331,
+      "eval_runtime": 59.3905,
+      "eval_samples_per_second": 93.315,
+      "eval_steps_per_second": 2.93,
       "step": 404
     },
     {
       "epoch": 4.06,
       "learning_rate": 2.02970297029703e-05,
-      "loss": 0.5553,
+      "loss": 0.2102,
       "step": 410
     },
     {
       "epoch": 4.16,
       "learning_rate": 2.079207920792079e-05,
-      "loss": 0.5075,
+      "loss": 0.1816,
       "step": 420
     },
     {
       "epoch": 4.26,
       "learning_rate": 2.128712871287129e-05,
-      "loss": 0.4948,
+      "loss": 0.195,
       "step": 430
     },
     {
       "epoch": 4.36,
       "learning_rate": 2.1782178217821783e-05,
-      "loss": 0.4892,
+      "loss": 0.1772,
       "step": 440
     },
     {
       "epoch": 4.45,
       "learning_rate": 2.227722772277228e-05,
-      "loss": 0.4861,
+      "loss": 0.2059,
       "step": 450
     },
     {
       "epoch": 4.55,
       "learning_rate": 2.2772277227722774e-05,
-      "loss": 0.4922,
+      "loss": 0.1992,
       "step": 460
     },
     {
       "epoch": 4.65,
       "learning_rate": 2.326732673267327e-05,
-      "loss": 0.498,
+      "loss": 0.2042,
       "step": 470
     },
     {
       "epoch": 4.75,
       "learning_rate": 2.3762376237623762e-05,
-      "loss": 0.4722,
+      "loss": 0.1945,
       "step": 480
     },
     {
       "epoch": 4.85,
       "learning_rate": 2.4257425742574257e-05,
-      "loss": 0.4619,
+      "loss": 0.1775,
       "step": 490
     },
     {
       "epoch": 4.95,
       "learning_rate": 2.4752475247524754e-05,
-      "loss": 0.4433,
+      "loss": 0.1395,
       "step": 500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.843558282208589,
-      "eval_loss": 0.4338299632072449,
-      "eval_runtime": 59.1883,
-      "eval_samples_per_second": 93.633,
-      "eval_steps_per_second": 2.94,
+      "eval_accuracy": 0.9133886683507759,
+      "eval_loss": 0.2648361623287201,
+      "eval_runtime": 59.3489,
+      "eval_samples_per_second": 93.38,
+      "eval_steps_per_second": 2.932,
       "step": 505
     },
     {
       "epoch": 5.05,
       "learning_rate": 2.5247524752475248e-05,
-      "loss": 0.4457,
+      "loss": 0.1543,
       "step": 510
     },
     {
       "epoch": 5.15,
       "learning_rate": 2.5742574257425746e-05,
-      "loss": 0.4402,
+      "loss": 0.1534,
       "step": 520
     },
     {
       "epoch": 5.25,
       "learning_rate": 2.623762376237624e-05,
-      "loss": 0.4118,
+      "loss": 0.1504,
       "step": 530
     },
     {
       "epoch": 5.35,
       "learning_rate": 2.6732673267326734e-05,
-      "loss": 0.4431,
+      "loss": 0.1909,
       "step": 540
     },
     {
       "epoch": 5.44,
       "learning_rate": 2.722772277227723e-05,
-      "loss": 0.4308,
+      "loss": 0.1579,
       "step": 550
     },
     {
       "epoch": 5.54,
       "learning_rate": 2.7722772277227726e-05,
-      "loss": 0.4262,
+      "loss": 0.1695,
       "step": 560
     },
     {
       "epoch": 5.64,
       "learning_rate": 2.8217821782178216e-05,
-      "loss": 0.4131,
+      "loss": 0.1638,
       "step": 570
     },
     {
       "epoch": 5.74,
       "learning_rate": 2.871287128712871e-05,
-      "loss": 0.3946,
+      "loss": 0.1574,
       "step": 580
     },
     {
       "epoch": 5.84,
       "learning_rate": 2.9207920792079208e-05,
-      "loss": 0.4015,
+      "loss": 0.1471,
       "step": 590
     },
     {
       "epoch": 5.94,
       "learning_rate": 2.9702970297029702e-05,
-      "loss": 0.406,
+      "loss": 0.155,
       "step": 600
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8661133164922411,
-      "eval_loss": 0.3765123784542084,
-      "eval_runtime": 59.3663,
-      "eval_samples_per_second": 93.353,
-      "eval_steps_per_second": 2.931,
+      "eval_accuracy": 0.915193071093468,
+      "eval_loss": 0.26563403010368347,
+      "eval_runtime": 59.8577,
+      "eval_samples_per_second": 92.586,
+      "eval_steps_per_second": 2.907,
       "step": 606
     },
     {
       "epoch": 6.04,
       "learning_rate": 3.01980198019802e-05,
-      "loss": 0.4215,
+      "loss": 0.1669,
       "step": 610
     },
     {
       "epoch": 6.14,
       "learning_rate": 3.06930693069307e-05,
-      "loss": 0.3587,
+      "loss": 0.137,
       "step": 620
     },
     {
       "epoch": 6.24,
       "learning_rate": 3.118811881188119e-05,
-      "loss": 0.3565,
+      "loss": 0.1461,
       "step": 630
     },
     {
       "epoch": 6.34,
       "learning_rate": 3.1683168316831686e-05,
-      "loss": 0.3848,
+      "loss": 0.1362,
       "step": 640
     },
     {
       "epoch": 6.43,
       "learning_rate": 3.217821782178218e-05,
-      "loss": 0.3508,
+      "loss": 0.1111,
       "step": 650
     },
     {
       "epoch": 6.53,
       "learning_rate": 3.2673267326732674e-05,
-      "loss": 0.3723,
+      "loss": 0.1515,
       "step": 660
     },
     {
       "epoch": 6.63,
       "learning_rate": 3.3168316831683175e-05,
-      "loss": 0.3643,
+      "loss": 0.1637,
       "step": 670
     },
     {
       "epoch": 6.73,
       "learning_rate": 3.366336633663367e-05,
-      "loss": 0.4062,
+      "loss": 0.1567,
       "step": 680
     },
     {
       "epoch": 6.83,
       "learning_rate": 3.415841584158416e-05,
-      "loss": 0.3767,
+      "loss": 0.1492,
       "step": 690
     },
     {
       "epoch": 6.93,
       "learning_rate": 3.465346534653465e-05,
-      "loss": 0.3517,
+      "loss": 0.1422,
       "step": 700
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.8792854565138939,
-      "eval_loss": 0.34659793972969055,
-      "eval_runtime": 59.2647,
-      "eval_samples_per_second": 93.513,
-      "eval_steps_per_second": 2.936,
+      "eval_accuracy": 0.9141104294478528,
+      "eval_loss": 0.26065394282341003,
+      "eval_runtime": 59.5203,
+      "eval_samples_per_second": 93.111,
+      "eval_steps_per_second": 2.923,
       "step": 707
     },
     {
       "epoch": 7.03,
       "learning_rate": 3.514851485148515e-05,
-      "loss": 0.3314,
+      "loss": 0.1479,
       "step": 710
     },
     {
       "epoch": 7.13,
       "learning_rate": 3.5643564356435645e-05,
-      "loss": 0.3319,
+      "loss": 0.1464,
       "step": 720
     },
     {
       "epoch": 7.23,
       "learning_rate": 3.613861386138614e-05,
-      "loss": 0.32,
+      "loss": 0.1469,
       "step": 730
     },
     {
       "epoch": 7.33,
       "learning_rate": 3.6633663366336634e-05,
-      "loss": 0.3448,
+      "loss": 0.1403,
       "step": 740
     },
     {
       "epoch": 7.42,
       "learning_rate": 3.712871287128713e-05,
-      "loss": 0.3242,
+      "loss": 0.1166,
       "step": 750
     },
     {
       "epoch": 7.52,
       "learning_rate": 3.762376237623763e-05,
-      "loss": 0.3241,
+      "loss": 0.1276,
       "step": 760
     },
     {
       "epoch": 7.62,
       "learning_rate": 3.811881188118812e-05,
-      "loss": 0.3379,
+      "loss": 0.1356,
       "step": 770
     },
     {
       "epoch": 7.72,
       "learning_rate": 3.861386138613862e-05,
-      "loss": 0.3392,
+      "loss": 0.1563,
       "step": 780
     },
     {
       "epoch": 7.82,
       "learning_rate": 3.910891089108911e-05,
-      "loss": 0.2918,
+      "loss": 0.1256,
       "step": 790
     },
     {
       "epoch": 7.92,
       "learning_rate": 3.9603960396039605e-05,
-      "loss": 0.312,
+      "loss": 0.1511,
       "step": 800
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8969686033922771,
-      "eval_loss": 0.3010924756526947,
-      "eval_runtime": 59.78,
-      "eval_samples_per_second": 92.707,
-      "eval_steps_per_second": 2.911,
+      "eval_accuracy": 0.9157343919162757,
+      "eval_loss": 0.25571852922439575,
+      "eval_runtime": 59.7308,
+      "eval_samples_per_second": 92.783,
+      "eval_steps_per_second": 2.913,
       "step": 808
     },
     {
       "epoch": 8.02,
       "learning_rate": 4.0099009900990106e-05,
-      "loss": 0.2918,
+      "loss": 0.165,
       "step": 810
     },
     {
       "epoch": 8.12,
       "learning_rate": 4.05940594059406e-05,
-      "loss": 0.308,
+      "loss": 0.2171,
       "step": 820
     },
     {
       "epoch": 8.22,
       "learning_rate": 4.108910891089109e-05,
-      "loss": 0.2897,
+      "loss": 0.2175,
       "step": 830
     },
     {
       "epoch": 8.32,
       "learning_rate": 4.158415841584158e-05,
-      "loss": 0.2688,
+      "loss": 0.1698,
       "step": 840
     },
     {
       "epoch": 8.41,
       "learning_rate": 4.207920792079208e-05,
-      "loss": 0.2767,
+      "loss": 0.183,
       "step": 850
     },
     {
       "epoch": 8.51,
       "learning_rate": 4.257425742574258e-05,
-      "loss": 0.2737,
+      "loss": 0.1738,
       "step": 860
     },
     {
       "epoch": 8.61,
       "learning_rate": 4.306930693069307e-05,
-      "loss": 0.2805,
+      "loss": 0.1784,
       "step": 870
     },
     {
       "epoch": 8.71,
       "learning_rate": 4.3564356435643565e-05,
-      "loss": 0.2799,
+      "loss": 0.1965,
       "step": 880
     },
     {
       "epoch": 8.81,
       "learning_rate": 4.405940594059406e-05,
-      "loss": 0.2634,
+      "loss": 0.2112,
       "step": 890
     },
     {
       "epoch": 8.91,
       "learning_rate": 4.455445544554456e-05,
-      "loss": 0.2842,
+      "loss": 0.1938,
       "step": 900
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.896066402020931,
-      "eval_loss": 0.2943052351474762,
-      "eval_runtime": 59.263,
-      "eval_samples_per_second": 93.515,
-      "eval_steps_per_second": 2.936,
+      "eval_accuracy": 0.9049079754601227,
+      "eval_loss": 0.2679148316383362,
+      "eval_runtime": 59.5803,
+      "eval_samples_per_second": 93.017,
+      "eval_steps_per_second": 2.92,
       "step": 909
     },
     {
-      "epoch": 9.0,
-      "step": 909,
-      "total_flos": 9.017904477803397e+18,
-      "train_loss": 0.6144383929350196,
-      "train_runtime": 3112.0693,
-      "train_samples_per_second": 415.479,
-      "train_steps_per_second": 3.245
+      "epoch": 9.01,
+      "learning_rate": 4.5049504950495054e-05,
+      "loss": 0.2518,
+      "step": 910
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 4.554455445544555e-05,
+      "loss": 0.1973,
+      "step": 920
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 4.603960396039604e-05,
+      "loss": 0.1798,
+      "step": 930
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 4.653465346534654e-05,
+      "loss": 0.1841,
+      "step": 940
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 4.702970297029703e-05,
+      "loss": 0.1806,
+      "step": 950
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 4.7524752475247525e-05,
+      "loss": 0.1602,
+      "step": 960
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 4.801980198019802e-05,
+      "loss": 0.1913,
+      "step": 970
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 4.851485148514851e-05,
+      "loss": 0.2061,
+      "step": 980
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 4.9009900990099014e-05,
+      "loss": 0.1838,
+      "step": 990
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 4.950495049504951e-05,
+      "loss": 0.1629,
+      "step": 1000
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 5e-05,
+      "loss": 0.2094,
+      "step": 1010
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.9137495488993144,
+      "eval_loss": 0.23915652930736542,
+      "eval_runtime": 59.6978,
+      "eval_samples_per_second": 92.834,
+      "eval_steps_per_second": 2.915,
+      "step": 1010
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 4.994499449944995e-05,
+      "loss": 0.2019,
+      "step": 1020
+    },
+    {
+      "epoch": 10.2,
+      "learning_rate": 4.9889988998899896e-05,
+      "loss": 0.1756,
+      "step": 1030
+    },
+    {
+      "epoch": 10.3,
+      "learning_rate": 4.9834983498349835e-05,
+      "loss": 0.1626,
+      "step": 1040
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 4.977997799779978e-05,
+      "loss": 0.1608,
+      "step": 1050
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 4.972497249724973e-05,
+      "loss": 0.1912,
+      "step": 1060
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 4.9669966996699675e-05,
+      "loss": 0.1716,
+      "step": 1070
+    },
+    {
+      "epoch": 10.69,
+      "learning_rate": 4.961496149614962e-05,
+      "loss": 0.1683,
+      "step": 1080
+    },
+    {
+      "epoch": 10.79,
+      "learning_rate": 4.955995599559956e-05,
+      "loss": 0.1497,
+      "step": 1090
+    },
+    {
+      "epoch": 10.89,
+      "learning_rate": 4.950495049504951e-05,
+      "loss": 0.1827,
+      "step": 1100
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 4.9449944994499455e-05,
+      "loss": 0.1835,
+      "step": 1110
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.9155539516420065,
+      "eval_loss": 0.23998872935771942,
+      "eval_runtime": 59.7253,
+      "eval_samples_per_second": 92.792,
+      "eval_steps_per_second": 2.913,
+      "step": 1111
+    },
+    {
+      "epoch": 11.0,
+      "step": 1111,
+      "total_flos": 1.1021883250648596e+19,
+      "train_loss": 0.1948466427711406,
+      "train_runtime": 3819.4144,
+      "train_samples_per_second": 338.534,
+      "train_steps_per_second": 2.644
     }
   ],
   "max_steps": 10100,
   "num_train_epochs": 100,
-  "total_flos": 9.017904477803397e+18,
+  "total_flos": 1.1021883250648596e+19,
   "trial_name": null,
   "trial_params": null
 }