End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 39.8,
-    "eval_accuracy": 1.0,
-    "eval_loss": 0.18712519109249115,
-    "eval_runtime": 3.4703,
-    "eval_samples_per_second": 14.696,
-    "eval_steps_per_second": 0.576,
-    "total_flos": 4.535730650653655e+17,
-    "train_loss": 0.6862462480862935,
-    "train_runtime": 1980.068,
-    "train_samples_per_second": 9.252,
-    "train_steps_per_second": 0.061
+    "eval_accuracy": 0.9814814814814815,
+    "eval_loss": 0.11687792092561722,
+    "eval_runtime": 3.9193,
+    "eval_samples_per_second": 13.778,
+    "eval_steps_per_second": 0.51,
+    "total_flos": 4.7393239534534656e+17,
+    "train_loss": 0.6489270607630412,
+    "train_runtime": 2182.6681,
+    "train_samples_per_second": 8.778,
+    "train_steps_per_second": 0.055
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 39.8,
-    "eval_accuracy": 1.0,
-    "eval_loss": 0.18712519109249115,
-    "eval_runtime": 3.4703,
-    "eval_samples_per_second": 14.696,
-    "eval_steps_per_second": 0.576
+    "eval_accuracy": 0.9814814814814815,
+    "eval_loss": 0.11687792092561722,
+    "eval_runtime": 3.9193,
+    "eval_samples_per_second": 13.778,
+    "eval_steps_per_second": 0.51
 }

runs/Nov30_12-18-33_4dbb018ffe5d/events.out.tfevents.1669813282.4dbb018ffe5d.78.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b2505a46774caa38c0c0feeb97c307a563fc809ecc16f4a1710475152ecc64b
+size 357

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 39.8,
-    "total_flos": 4.535730650653655e+17,
-    "train_loss": 0.6862462480862935,
-    "train_runtime": 1980.068,
-    "train_samples_per_second": 9.252,
-    "train_steps_per_second": 0.061
+    "total_flos": 4.7393239534534656e+17,
+    "train_loss": 0.6489270607630412,
+    "train_runtime": 2182.6681,
+    "train_samples_per_second": 8.778,
+    "train_steps_per_second": 0.055
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 1.0,
-  "best_model_checkpoint": "delivery_truck_classification/checkpoint-33",
+  "best_metric": 0.9814814814814815,
+  "best_model_checkpoint": "delivery_truck_classification/checkpoint-81",
   "epoch": 39.8,
   "global_step": 120,
   "is_hyper_param_search": false,
@@ -9,413 +9,413 @@
   "log_history": [
     {
       "epoch": 0.8,
-      "eval_accuracy": 0.0392156862745098,
-      "eval_loss": 1.8911739587783813,
-      "eval_runtime": 3.471,
-      "eval_samples_per_second": 14.693,
-      "eval_steps_per_second": 0.576,
+      "eval_accuracy": 0.2037037037037037,
+      "eval_loss": 1.7556031942367554,
+      "eval_runtime": 3.778,
+      "eval_samples_per_second": 14.293,
+      "eval_steps_per_second": 0.529,
       "step": 3
     },
     {
       "epoch": 1.8,
-      "eval_accuracy": 0.27450980392156865,
-      "eval_loss": 1.7519241571426392,
-      "eval_runtime": 3.9826,
-      "eval_samples_per_second": 12.806,
-      "eval_steps_per_second": 0.502,
+      "eval_accuracy": 0.37037037037037035,
+      "eval_loss": 1.5833455324172974,
+      "eval_runtime": 4.1987,
+      "eval_samples_per_second": 12.861,
+      "eval_steps_per_second": 0.476,
       "step": 6
     },
     {
       "epoch": 2.8,
-      "eval_accuracy": 0.47058823529411764,
-      "eval_loss": 1.5549100637435913,
-      "eval_runtime": 3.7484,
-      "eval_samples_per_second": 13.606,
-      "eval_steps_per_second": 0.534,
+      "eval_accuracy": 0.5925925925925926,
+      "eval_loss": 1.3483160734176636,
+      "eval_runtime": 3.9811,
+      "eval_samples_per_second": 13.564,
+      "eval_steps_per_second": 0.502,
       "step": 9
     },
     {
       "epoch": 3.8,
       "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 1.2850992679595947,
-      "eval_runtime": 3.3946,
-      "eval_samples_per_second": 15.024,
-      "eval_steps_per_second": 0.589,
+      "eval_loss": 1.1101388931274414,
+      "eval_runtime": 3.792,
+      "eval_samples_per_second": 14.241,
+      "eval_steps_per_second": 0.527,
       "step": 12
     },
     {
       "epoch": 4.8,
-      "eval_accuracy": 0.7647058823529411,
-      "eval_loss": 0.9967808723449707,
-      "eval_runtime": 3.7653,
-      "eval_samples_per_second": 13.545,
-      "eval_steps_per_second": 0.531,
+      "eval_accuracy": 0.7222222222222222,
+      "eval_loss": 0.9116207957267761,
+      "eval_runtime": 4.055,
+      "eval_samples_per_second": 13.317,
+      "eval_steps_per_second": 0.493,
       "step": 15
     },
     {
       "epoch": 5.8,
-      "eval_accuracy": 0.7843137254901961,
-      "eval_loss": 0.7825759649276733,
-      "eval_runtime": 3.7762,
-      "eval_samples_per_second": 13.506,
-      "eval_steps_per_second": 0.53,
+      "eval_accuracy": 0.7407407407407407,
+      "eval_loss": 0.7631537914276123,
+      "eval_runtime": 4.0162,
+      "eval_samples_per_second": 13.446,
+      "eval_steps_per_second": 0.498,
       "step": 18
     },
     {
       "epoch": 6.53,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 1.787,
+      "loss": 1.7322,
       "step": 20
     },
     {
       "epoch": 6.8,
-      "eval_accuracy": 0.8823529411764706,
-      "eval_loss": 0.600950300693512,
-      "eval_runtime": 3.4707,
-      "eval_samples_per_second": 14.694,
-      "eval_steps_per_second": 0.576,
+      "eval_accuracy": 0.7962962962962963,
+      "eval_loss": 0.6118068695068359,
+      "eval_runtime": 3.8828,
+      "eval_samples_per_second": 13.908,
+      "eval_steps_per_second": 0.515,
       "step": 21
     },
     {
       "epoch": 7.8,
-      "eval_accuracy": 0.9019607843137255,
-      "eval_loss": 0.43014010787010193,
-      "eval_runtime": 3.7555,
-      "eval_samples_per_second": 13.58,
-      "eval_steps_per_second": 0.533,
+      "eval_accuracy": 0.8518518518518519,
+      "eval_loss": 0.501708984375,
+      "eval_runtime": 4.0233,
+      "eval_samples_per_second": 13.422,
+      "eval_steps_per_second": 0.497,
       "step": 24
     },
     {
       "epoch": 8.8,
-      "eval_accuracy": 0.8823529411764706,
-      "eval_loss": 0.3232599198818207,
-      "eval_runtime": 3.738,
-      "eval_samples_per_second": 13.644,
-      "eval_steps_per_second": 0.535,
+      "eval_accuracy": 0.8888888888888888,
+      "eval_loss": 0.4240618944168091,
+      "eval_runtime": 4.0079,
+      "eval_samples_per_second": 13.473,
+      "eval_steps_per_second": 0.499,
       "step": 27
     },
     {
       "epoch": 9.8,
-      "eval_accuracy": 0.9411764705882353,
-      "eval_loss": 0.2302764505147934,
-      "eval_runtime": 3.4208,
-      "eval_samples_per_second": 14.909,
-      "eval_steps_per_second": 0.585,
+      "eval_accuracy": 0.8703703703703703,
+      "eval_loss": 0.3522408902645111,
+      "eval_runtime": 3.8114,
+      "eval_samples_per_second": 14.168,
+      "eval_steps_per_second": 0.525,
       "step": 30
     },
     {
       "epoch": 10.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.18712519109249115,
-      "eval_runtime": 3.7332,
-      "eval_samples_per_second": 13.661,
-      "eval_steps_per_second": 0.536,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.2918209135532379,
+      "eval_runtime": 4.0352,
+      "eval_samples_per_second": 13.382,
+      "eval_steps_per_second": 0.496,
       "step": 33
     },
     {
       "epoch": 11.8,
-      "eval_accuracy": 0.9607843137254902,
-      "eval_loss": 0.16001677513122559,
-      "eval_runtime": 3.7595,
-      "eval_samples_per_second": 13.566,
-      "eval_steps_per_second": 0.532,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.2659091353416443,
+      "eval_runtime": 4.0341,
+      "eval_samples_per_second": 13.386,
+      "eval_steps_per_second": 0.496,
       "step": 36
     },
     {
       "epoch": 12.8,
-      "eval_accuracy": 0.9803921568627451,
-      "eval_loss": 0.13339409232139587,
-      "eval_runtime": 3.4512,
-      "eval_samples_per_second": 14.778,
-      "eval_steps_per_second": 0.58,
+      "eval_accuracy": 0.9444444444444444,
+      "eval_loss": 0.2587053179740906,
+      "eval_runtime": 3.9977,
+      "eval_samples_per_second": 13.508,
+      "eval_steps_per_second": 0.5,
       "step": 39
     },
     {
       "epoch": 13.27,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.7554,
+      "loss": 0.7462,
       "step": 40
     },
     {
       "epoch": 13.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.10250363498926163,
-      "eval_runtime": 3.7508,
-      "eval_samples_per_second": 13.597,
-      "eval_steps_per_second": 0.533,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.20625640451908112,
+      "eval_runtime": 3.9968,
+      "eval_samples_per_second": 13.511,
+      "eval_steps_per_second": 0.5,
       "step": 42
     },
     {
       "epoch": 14.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.09086543321609497,
-      "eval_runtime": 3.7591,
-      "eval_samples_per_second": 13.567,
-      "eval_steps_per_second": 0.532,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.1870189756155014,
+      "eval_runtime": 4.0713,
+      "eval_samples_per_second": 13.264,
+      "eval_steps_per_second": 0.491,
       "step": 45
     },
     {
       "epoch": 15.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.07331494987010956,
-      "eval_runtime": 3.4513,
-      "eval_samples_per_second": 14.777,
-      "eval_steps_per_second": 0.579,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.1739373356103897,
+      "eval_runtime": 3.7563,
+      "eval_samples_per_second": 14.376,
+      "eval_steps_per_second": 0.532,
       "step": 48
     },
     {
       "epoch": 16.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.062474265694618225,
-      "eval_runtime": 3.8017,
-      "eval_samples_per_second": 13.415,
-      "eval_steps_per_second": 0.526,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.20426180958747864,
+      "eval_runtime": 4.0894,
+      "eval_samples_per_second": 13.205,
+      "eval_steps_per_second": 0.489,
       "step": 51
     },
     {
       "epoch": 17.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.060241490602493286,
-      "eval_runtime": 3.8432,
-      "eval_samples_per_second": 13.27,
-      "eval_steps_per_second": 0.52,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.18968288600444794,
+      "eval_runtime": 4.1068,
+      "eval_samples_per_second": 13.149,
+      "eval_steps_per_second": 0.487,
       "step": 54
     },
     {
       "epoch": 18.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.061301086097955704,
-      "eval_runtime": 3.4257,
-      "eval_samples_per_second": 14.888,
-      "eval_steps_per_second": 0.584,
+      "eval_accuracy": 0.9444444444444444,
+      "eval_loss": 0.17641673982143402,
+      "eval_runtime": 3.7856,
+      "eval_samples_per_second": 14.265,
+      "eval_steps_per_second": 0.528,
       "step": 57
     },
     {
       "epoch": 19.8,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.4731,
+      "loss": 0.4232,
       "step": 60
     },
     {
       "epoch": 19.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.0506136529147625,
-      "eval_runtime": 3.7258,
-      "eval_samples_per_second": 13.688,
-      "eval_steps_per_second": 0.537,
+      "eval_accuracy": 0.9444444444444444,
+      "eval_loss": 0.1586901694536209,
+      "eval_runtime": 4.4136,
+      "eval_samples_per_second": 12.235,
+      "eval_steps_per_second": 0.453,
       "step": 60
     },
     {
       "epoch": 20.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.05875474214553833,
-      "eval_runtime": 3.7701,
-      "eval_samples_per_second": 13.528,
-      "eval_steps_per_second": 0.53,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.15556305646896362,
+      "eval_runtime": 4.0371,
+      "eval_samples_per_second": 13.376,
+      "eval_steps_per_second": 0.495,
       "step": 63
     },
     {
       "epoch": 21.8,
-      "eval_accuracy": 0.9803921568627451,
-      "eval_loss": 0.06549996882677078,
-      "eval_runtime": 3.4174,
-      "eval_samples_per_second": 14.924,
-      "eval_steps_per_second": 0.585,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.15161330997943878,
+      "eval_runtime": 3.8866,
+      "eval_samples_per_second": 13.894,
+      "eval_steps_per_second": 0.515,
       "step": 66
     },
     {
       "epoch": 22.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.05172449350357056,
-      "eval_runtime": 3.8113,
-      "eval_samples_per_second": 13.381,
-      "eval_steps_per_second": 0.525,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.12640085816383362,
+      "eval_runtime": 3.9739,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 0.503,
       "step": 69
     },
     {
       "epoch": 23.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.04140729829668999,
-      "eval_runtime": 3.8035,
-      "eval_samples_per_second": 13.409,
-      "eval_steps_per_second": 0.526,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.11799591779708862,
+      "eval_runtime": 4.1809,
+      "eval_samples_per_second": 12.916,
+      "eval_steps_per_second": 0.478,
       "step": 72
     },
     {
       "epoch": 24.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.04080395773053169,
-      "eval_runtime": 3.4453,
-      "eval_samples_per_second": 14.803,
-      "eval_steps_per_second": 0.581,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.11095747351646423,
+      "eval_runtime": 3.7145,
+      "eval_samples_per_second": 14.537,
+      "eval_steps_per_second": 0.538,
       "step": 75
     },
     {
       "epoch": 25.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.041679248213768005,
-      "eval_runtime": 3.7337,
-      "eval_samples_per_second": 13.66,
-      "eval_steps_per_second": 0.536,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.12324906885623932,
+      "eval_runtime": 4.1986,
+      "eval_samples_per_second": 12.861,
+      "eval_steps_per_second": 0.476,
       "step": 78
     },
     {
       "epoch": 26.53,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.4248,
+      "loss": 0.3571,
       "step": 80
     },
     {
       "epoch": 26.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.03893385827541351,
-      "eval_runtime": 3.8193,
-      "eval_samples_per_second": 13.353,
-      "eval_steps_per_second": 0.524,
+      "eval_accuracy": 0.9814814814814815,
+      "eval_loss": 0.11687792092561722,
+      "eval_runtime": 4.0469,
+      "eval_samples_per_second": 13.344,
+      "eval_steps_per_second": 0.494,
       "step": 81
     },
     {
       "epoch": 27.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.03756618872284889,
-      "eval_runtime": 3.4098,
-      "eval_samples_per_second": 14.957,
-      "eval_steps_per_second": 0.587,
+      "eval_accuracy": 0.9814814814814815,
+      "eval_loss": 0.10509185492992401,
+      "eval_runtime": 3.8141,
+      "eval_samples_per_second": 14.158,
+      "eval_steps_per_second": 0.524,
       "step": 84
     },
     {
       "epoch": 28.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.036111168563365936,
-      "eval_runtime": 3.7255,
-      "eval_samples_per_second": 13.689,
-      "eval_steps_per_second": 0.537,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.09857716411352158,
+      "eval_runtime": 4.0141,
+      "eval_samples_per_second": 13.452,
+      "eval_steps_per_second": 0.498,
       "step": 87
     },
     {
       "epoch": 29.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.035053811967372894,
-      "eval_runtime": 3.7896,
-      "eval_samples_per_second": 13.458,
-      "eval_steps_per_second": 0.528,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.09369004517793655,
+      "eval_runtime": 4.0266,
+      "eval_samples_per_second": 13.411,
+      "eval_steps_per_second": 0.497,
       "step": 90
     },
     {
       "epoch": 30.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.029912708327174187,
-      "eval_runtime": 3.447,
-      "eval_samples_per_second": 14.796,
-      "eval_steps_per_second": 0.58,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.09307903051376343,
+      "eval_runtime": 3.8076,
+      "eval_samples_per_second": 14.182,
+      "eval_steps_per_second": 0.525,
       "step": 93
     },
     {
       "epoch": 31.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.028354115784168243,
-      "eval_runtime": 3.8199,
-      "eval_samples_per_second": 13.351,
-      "eval_steps_per_second": 0.524,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.09316574782133102,
+      "eval_runtime": 3.9741,
+      "eval_samples_per_second": 13.588,
+      "eval_steps_per_second": 0.503,
       "step": 96
     },
     {
       "epoch": 32.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.027853690087795258,
-      "eval_runtime": 3.7966,
-      "eval_samples_per_second": 13.433,
-      "eval_steps_per_second": 0.527,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.0940571129322052,
+      "eval_runtime": 4.072,
+      "eval_samples_per_second": 13.261,
+      "eval_steps_per_second": 0.491,
       "step": 99
     },
     {
       "epoch": 33.27,
       "learning_rate": 9.259259259259259e-06,
-      "loss": 0.3657,
+      "loss": 0.3239,
       "step": 100
     },
     {
       "epoch": 33.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.027496235445141792,
-      "eval_runtime": 3.4451,
-      "eval_samples_per_second": 14.804,
-      "eval_steps_per_second": 0.581,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.09196292608976364,
+      "eval_runtime": 3.764,
+      "eval_samples_per_second": 14.347,
+      "eval_steps_per_second": 0.531,
       "step": 102
     },
     {
       "epoch": 34.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.027940476313233376,
-      "eval_runtime": 3.7579,
-      "eval_samples_per_second": 13.571,
-      "eval_steps_per_second": 0.532,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.08508706837892532,
+      "eval_runtime": 3.9778,
+      "eval_samples_per_second": 13.575,
+      "eval_steps_per_second": 0.503,
       "step": 105
     },
     {
       "epoch": 35.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.027937183156609535,
-      "eval_runtime": 3.8018,
-      "eval_samples_per_second": 13.415,
-      "eval_steps_per_second": 0.526,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.08276030421257019,
+      "eval_runtime": 4.0417,
+      "eval_samples_per_second": 13.361,
+      "eval_steps_per_second": 0.495,
       "step": 108
     },
     {
       "epoch": 36.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.02781843952834606,
-      "eval_runtime": 3.4196,
-      "eval_samples_per_second": 14.914,
-      "eval_steps_per_second": 0.585,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.0810302197933197,
+      "eval_runtime": 3.7281,
+      "eval_samples_per_second": 14.485,
+      "eval_steps_per_second": 0.536,
       "step": 111
     },
     {
       "epoch": 37.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.027574485167860985,
-      "eval_runtime": 3.7687,
-      "eval_samples_per_second": 13.532,
-      "eval_steps_per_second": 0.531,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.0800759419798851,
+      "eval_runtime": 4.1175,
+      "eval_samples_per_second": 13.115,
+      "eval_steps_per_second": 0.486,
       "step": 114
     },
     {
       "epoch": 38.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.02744271047413349,
-      "eval_runtime": 3.7748,
-      "eval_samples_per_second": 13.511,
-      "eval_steps_per_second": 0.53,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.08043589442968369,
+      "eval_runtime": 3.9928,
+      "eval_samples_per_second": 13.524,
+      "eval_steps_per_second": 0.501,
       "step": 117
     },
     {
       "epoch": 39.8,
       "learning_rate": 0.0,
-      "loss": 0.3115,
+      "loss": 0.3111,
       "step": 120
     },
     {
       "epoch": 39.8,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.02739746682345867,
-      "eval_runtime": 3.4498,
-      "eval_samples_per_second": 14.784,
-      "eval_steps_per_second": 0.58,
+      "eval_accuracy": 0.9629629629629629,
+      "eval_loss": 0.08069680631160736,
+      "eval_runtime": 3.8487,
+      "eval_samples_per_second": 14.031,
+      "eval_steps_per_second": 0.52,
       "step": 120
     },
     {
       "epoch": 39.8,
       "step": 120,
-      "total_flos": 4.535730650653655e+17,
-      "train_loss": 0.6862462480862935,
-      "train_runtime": 1980.068,
-      "train_samples_per_second": 9.252,
-      "train_steps_per_second": 0.061
+      "total_flos": 4.7393239534534656e+17,
+      "train_loss": 0.6489270607630412,
+      "train_runtime": 2182.6681,
+      "train_samples_per_second": 8.778,
+      "train_steps_per_second": 0.055
     }
   ],
   "max_steps": 120,
   "num_train_epochs": 40,
-  "total_flos": 4.535730650653655e+17,
+  "total_flos": 4.7393239534534656e+17,
   "trial_name": null,
   "trial_params": null
 }