End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 39.8,
-    "eval_accuracy": 0.9814814814814815,
-    "eval_loss": 0.11687792092561722,
-    "eval_runtime": 3.9193,
-    "eval_samples_per_second": 13.778,
-    "eval_steps_per_second": 0.51,
+    "eval_accuracy": 0.9259259259259259,
+    "eval_loss": 0.24187135696411133,
+    "eval_runtime": 3.8793,
+    "eval_samples_per_second": 13.92,
+    "eval_steps_per_second": 0.516,
     "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
+    "train_loss": 0.6987951993942261,
+    "train_runtime": 2118.2023,
+    "train_samples_per_second": 9.045,
+    "train_steps_per_second": 0.057
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 39.8,
-    "eval_accuracy": 0.9814814814814815,
-    "eval_loss": 0.11687792092561722,
-    "eval_runtime": 3.9193,
-    "eval_samples_per_second": 13.778,
-    "eval_steps_per_second": 0.51
+    "eval_accuracy": 0.9259259259259259,
+    "eval_loss": 0.24187135696411133,
+    "eval_runtime": 3.8793,
+    "eval_samples_per_second": 13.92,
+    "eval_steps_per_second": 0.516
 }

runs/Nov30_17-36-36_b2664fd97c09/events.out.tfevents.1669832223.b2664fd97c09.78.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b3cab9c2ea0631d309ac899312df84867398882770ca459eac126133f3dc511
+size 357

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 39.8,
     "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
+    "train_loss": 0.6987951993942261,
+    "train_runtime": 2118.2023,
+    "train_samples_per_second": 9.045,
+    "train_steps_per_second": 0.057
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9814814814814815,
-  "best_model_checkpoint": "delivery_truck_classification/checkpoint-81",
+  "best_metric": 0.9259259259259259,
+  "best_model_checkpoint": "delivery_truck_classification/checkpoint-66",
   "epoch": 39.8,
   "global_step": 120,
   "is_hyper_param_search": false,
@@ -9,408 +9,408 @@
   "log_history": [
     {
       "epoch": 0.8,
-      "eval_accuracy": 0.2037037037037037,
-      "eval_loss": 1.7556031942367554,
-      "eval_runtime": 3.778,
-      "eval_samples_per_second": 14.293,
-      "eval_steps_per_second": 0.529,
+      "eval_accuracy": 0.2222222222222222,
+      "eval_loss": 1.8672751188278198,
+      "eval_runtime": 3.8871,
+      "eval_samples_per_second": 13.892,
+      "eval_steps_per_second": 0.515,
       "step": 3
     },
     {
       "epoch": 1.8,
-      "eval_accuracy": 0.37037037037037035,
-      "eval_loss": 1.5833455324172974,
-      "eval_runtime": 4.1987,
-      "eval_samples_per_second": 12.861,
-      "eval_steps_per_second": 0.476,
+      "eval_accuracy": 0.25925925925925924,
+      "eval_loss": 1.7420549392700195,
+      "eval_runtime": 4.3749,
+      "eval_samples_per_second": 12.343,
+      "eval_steps_per_second": 0.457,
       "step": 6
     },
     {
       "epoch": 2.8,
-      "eval_accuracy": 0.5925925925925926,
-      "eval_loss": 1.3483160734176636,
-      "eval_runtime": 3.9811,
-      "eval_samples_per_second": 13.564,
-      "eval_steps_per_second": 0.502,
+      "eval_accuracy": 0.42592592592592593,
+      "eval_loss": 1.5909960269927979,
+      "eval_runtime": 4.1131,
+      "eval_samples_per_second": 13.129,
+      "eval_steps_per_second": 0.486,
       "step": 9
     },
     {
       "epoch": 3.8,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 1.1101388931274414,
-      "eval_runtime": 3.792,
-      "eval_samples_per_second": 14.241,
-      "eval_steps_per_second": 0.527,
+      "eval_accuracy": 0.5,
+      "eval_loss": 1.4371126890182495,
+      "eval_runtime": 3.8435,
+      "eval_samples_per_second": 14.05,
+      "eval_steps_per_second": 0.52,
       "step": 12
     },
     {
       "epoch": 4.8,
-      "eval_accuracy": 0.7222222222222222,
-      "eval_loss": 0.9116207957267761,
-      "eval_runtime": 4.055,
-      "eval_samples_per_second": 13.317,
-      "eval_steps_per_second": 0.493,
+      "eval_accuracy": 0.5740740740740741,
+      "eval_loss": 1.2870995998382568,
+      "eval_runtime": 4.1635,
+      "eval_samples_per_second": 12.97,
+      "eval_steps_per_second": 0.48,
       "step": 15
     },
     {
       "epoch": 5.8,
-      "eval_accuracy": 0.7407407407407407,
-      "eval_loss": 0.7631537914276123,
-      "eval_runtime": 4.0162,
-      "eval_samples_per_second": 13.446,
-      "eval_steps_per_second": 0.498,
+      "eval_accuracy": 0.5740740740740741,
+      "eval_loss": 1.1510515213012695,
+      "eval_runtime": 4.0675,
+      "eval_samples_per_second": 13.276,
+      "eval_steps_per_second": 0.492,
       "step": 18
     },
     {
       "epoch": 6.53,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 1.7322,
+      "loss": 1.8164,
       "step": 20
     },
     {
       "epoch": 6.8,
-      "eval_accuracy": 0.7962962962962963,
-      "eval_loss": 0.6118068695068359,
-      "eval_runtime": 3.8828,
-      "eval_samples_per_second": 13.908,
-      "eval_steps_per_second": 0.515,
+      "eval_accuracy": 0.7222222222222222,
+      "eval_loss": 0.9362860321998596,
+      "eval_runtime": 3.7257,
+      "eval_samples_per_second": 14.494,
+      "eval_steps_per_second": 0.537,
       "step": 21
     },
     {
       "epoch": 7.8,
-      "eval_accuracy": 0.8518518518518519,
-      "eval_loss": 0.501708984375,
-      "eval_runtime": 4.0233,
-      "eval_samples_per_second": 13.422,
-      "eval_steps_per_second": 0.497,
+      "eval_accuracy": 0.7777777777777778,
+      "eval_loss": 0.7902935147285461,
+      "eval_runtime": 4.0908,
+      "eval_samples_per_second": 13.2,
+      "eval_steps_per_second": 0.489,
       "step": 24
     },
     {
       "epoch": 8.8,
-      "eval_accuracy": 0.8888888888888888,
-      "eval_loss": 0.4240618944168091,
-      "eval_runtime": 4.0079,
-      "eval_samples_per_second": 13.473,
-      "eval_steps_per_second": 0.499,
+      "eval_accuracy": 0.7592592592592593,
+      "eval_loss": 0.6839045882225037,
+      "eval_runtime": 4.2293,
+      "eval_samples_per_second": 12.768,
+      "eval_steps_per_second": 0.473,
       "step": 27
     },
     {
       "epoch": 9.8,
-      "eval_accuracy": 0.8703703703703703,
-      "eval_loss": 0.3522408902645111,
-      "eval_runtime": 3.8114,
-      "eval_samples_per_second": 14.168,
-      "eval_steps_per_second": 0.525,
+      "eval_accuracy": 0.7777777777777778,
+      "eval_loss": 0.5660675764083862,
+      "eval_runtime": 3.7011,
+      "eval_samples_per_second": 14.59,
+      "eval_steps_per_second": 0.54,
       "step": 30
     },
     {
       "epoch": 10.8,
-      "eval_accuracy": 0.9259259259259259,
-      "eval_loss": 0.2918209135532379,
-      "eval_runtime": 4.0352,
-      "eval_samples_per_second": 13.382,
-      "eval_steps_per_second": 0.496,
+      "eval_accuracy": 0.8518518518518519,
+      "eval_loss": 0.4638045132160187,
+      "eval_runtime": 4.0924,
+      "eval_samples_per_second": 13.195,
+      "eval_steps_per_second": 0.489,
       "step": 33
     },
     {
       "epoch": 11.8,
-      "eval_accuracy": 0.9259259259259259,
-      "eval_loss": 0.2659091353416443,
-      "eval_runtime": 4.0341,
-      "eval_samples_per_second": 13.386,
-      "eval_steps_per_second": 0.496,
+      "eval_accuracy": 0.8703703703703703,
+      "eval_loss": 0.4015035033226013,
+      "eval_runtime": 4.126,
+      "eval_samples_per_second": 13.088,
+      "eval_steps_per_second": 0.485,
       "step": 36
     },
     {
       "epoch": 12.8,
-      "eval_accuracy": 0.9444444444444444,
-      "eval_loss": 0.2587053179740906,
-      "eval_runtime": 3.9977,
-      "eval_samples_per_second": 13.508,
-      "eval_steps_per_second": 0.5,
+      "eval_accuracy": 0.8703703703703703,
+      "eval_loss": 0.3809252083301544,
+      "eval_runtime": 3.7332,
+      "eval_samples_per_second": 14.465,
+      "eval_steps_per_second": 0.536,
       "step": 39
     },
     {
       "epoch": 13.27,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.7462,
+      "loss": 0.8525,
       "step": 40
     },
     {
       "epoch": 13.8,
-      "eval_accuracy": 0.9259259259259259,
-      "eval_loss": 0.20625640451908112,
-      "eval_runtime": 3.9968,
-      "eval_samples_per_second": 13.511,
-      "eval_steps_per_second": 0.5,
+      "eval_accuracy": 0.9074074074074074,
+      "eval_loss": 0.32140079140663147,
+      "eval_runtime": 4.0811,
+      "eval_samples_per_second": 13.232,
+      "eval_steps_per_second": 0.49,
       "step": 42
     },
     {
       "epoch": 14.8,
-      "eval_accuracy": 0.9259259259259259,
-      "eval_loss": 0.1870189756155014,
-      "eval_runtime": 4.0713,
-      "eval_samples_per_second": 13.264,
-      "eval_steps_per_second": 0.491,
+      "eval_accuracy": 0.8703703703703703,
+      "eval_loss": 0.31137827038764954,
+      "eval_runtime": 4.0629,
+      "eval_samples_per_second": 13.291,
+      "eval_steps_per_second": 0.492,
       "step": 45
     },
     {
       "epoch": 15.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.1739373356103897,
-      "eval_runtime": 3.7563,
-      "eval_samples_per_second": 14.376,
-      "eval_steps_per_second": 0.532,
+      "eval_accuracy": 0.8888888888888888,
+      "eval_loss": 0.30256393551826477,
+      "eval_runtime": 3.8458,
+      "eval_samples_per_second": 14.041,
+      "eval_steps_per_second": 0.52,
       "step": 48
     },
     {
       "epoch": 16.8,
-      "eval_accuracy": 0.9259259259259259,
-      "eval_loss": 0.20426180958747864,
-      "eval_runtime": 4.0894,
-      "eval_samples_per_second": 13.205,
+      "eval_accuracy": 0.8888888888888888,
+      "eval_loss": 0.2969690263271332,
+      "eval_runtime": 4.091,
+      "eval_samples_per_second": 13.2,
       "eval_steps_per_second": 0.489,
       "step": 51
     },
     {
       "epoch": 17.8,
-      "eval_accuracy": 0.9259259259259259,
-      "eval_loss": 0.18968288600444794,
-      "eval_runtime": 4.1068,
-      "eval_samples_per_second": 13.149,
-      "eval_steps_per_second": 0.487,
+      "eval_accuracy": 0.8888888888888888,
+      "eval_loss": 0.2597466707229614,
+      "eval_runtime": 4.0172,
+      "eval_samples_per_second": 13.442,
+      "eval_steps_per_second": 0.498,
       "step": 54
     },
     {
       "epoch": 18.8,
-      "eval_accuracy": 0.9444444444444444,
-      "eval_loss": 0.17641673982143402,
-      "eval_runtime": 3.7856,
-      "eval_samples_per_second": 14.265,
-      "eval_steps_per_second": 0.528,
+      "eval_accuracy": 0.8888888888888888,
+      "eval_loss": 0.2791738212108612,
+      "eval_runtime": 3.7642,
+      "eval_samples_per_second": 14.346,
+      "eval_steps_per_second": 0.531,
       "step": 57
     },
     {
       "epoch": 19.8,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.4232,
+      "loss": 0.4831,
       "step": 60
     },
     {
       "epoch": 19.8,
-      "eval_accuracy": 0.9444444444444444,
-      "eval_loss": 0.1586901694536209,
-      "eval_runtime": 4.4136,
-      "eval_samples_per_second": 12.235,
-      "eval_steps_per_second": 0.453,
+      "eval_accuracy": 0.8703703703703703,
+      "eval_loss": 0.3208976984024048,
+      "eval_runtime": 4.0804,
+      "eval_samples_per_second": 13.234,
+      "eval_steps_per_second": 0.49,
       "step": 60
     },
     {
       "epoch": 20.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.15556305646896362,
-      "eval_runtime": 4.0371,
-      "eval_samples_per_second": 13.376,
-      "eval_steps_per_second": 0.495,
+      "eval_accuracy": 0.9074074074074074,
+      "eval_loss": 0.2928622364997864,
+      "eval_runtime": 4.1417,
+      "eval_samples_per_second": 13.038,
+      "eval_steps_per_second": 0.483,
       "step": 63
     },
     {
       "epoch": 21.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.15161330997943878,
-      "eval_runtime": 3.8866,
-      "eval_samples_per_second": 13.894,
-      "eval_steps_per_second": 0.515,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.24187135696411133,
+      "eval_runtime": 3.7664,
+      "eval_samples_per_second": 14.337,
+      "eval_steps_per_second": 0.531,
       "step": 66
     },
     {
       "epoch": 22.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.12640085816383362,
-      "eval_runtime": 3.9739,
-      "eval_samples_per_second": 13.589,
-      "eval_steps_per_second": 0.503,
+      "eval_accuracy": 0.9074074074074074,
+      "eval_loss": 0.24963845312595367,
+      "eval_runtime": 4.1506,
+      "eval_samples_per_second": 13.01,
+      "eval_steps_per_second": 0.482,
       "step": 69
     },
     {
       "epoch": 23.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.11799591779708862,
-      "eval_runtime": 4.1809,
-      "eval_samples_per_second": 12.916,
+      "eval_accuracy": 0.9074074074074074,
+      "eval_loss": 0.29534202814102173,
+      "eval_runtime": 4.1866,
+      "eval_samples_per_second": 12.898,
       "eval_steps_per_second": 0.478,
       "step": 72
     },
     {
       "epoch": 24.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.11095747351646423,
-      "eval_runtime": 3.7145,
-      "eval_samples_per_second": 14.537,
-      "eval_steps_per_second": 0.538,
+      "eval_accuracy": 0.8888888888888888,
+      "eval_loss": 0.30937835574150085,
+      "eval_runtime": 3.7669,
+      "eval_samples_per_second": 14.336,
+      "eval_steps_per_second": 0.531,
       "step": 75
     },
     {
       "epoch": 25.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.12324906885623932,
-      "eval_runtime": 4.1986,
-      "eval_samples_per_second": 12.861,
-      "eval_steps_per_second": 0.476,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.2792222201824188,
+      "eval_runtime": 4.0339,
+      "eval_samples_per_second": 13.387,
+      "eval_steps_per_second": 0.496,
       "step": 78
     },
     {
       "epoch": 26.53,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.3571,
+      "loss": 0.3889,
       "step": 80
     },
     {
       "epoch": 26.8,
-      "eval_accuracy": 0.9814814814814815,
-      "eval_loss": 0.11687792092561722,
-      "eval_runtime": 4.0469,
-      "eval_samples_per_second": 13.344,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.2522352933883667,
+      "eval_runtime": 4.0446,
+      "eval_samples_per_second": 13.351,
       "eval_steps_per_second": 0.494,
       "step": 81
     },
     {
       "epoch": 27.8,
-      "eval_accuracy": 0.9814814814814815,
-      "eval_loss": 0.10509185492992401,
-      "eval_runtime": 3.8141,
-      "eval_samples_per_second": 14.158,
-      "eval_steps_per_second": 0.524,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.24507637321949005,
+      "eval_runtime": 3.701,
+      "eval_samples_per_second": 14.591,
+      "eval_steps_per_second": 0.54,
       "step": 84
     },
     {
       "epoch": 28.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.09857716411352158,
-      "eval_runtime": 4.0141,
-      "eval_samples_per_second": 13.452,
-      "eval_steps_per_second": 0.498,
+      "eval_accuracy": 0.9074074074074074,
+      "eval_loss": 0.2540915906429291,
+      "eval_runtime": 4.1415,
+      "eval_samples_per_second": 13.039,
+      "eval_steps_per_second": 0.483,
       "step": 87
     },
     {
       "epoch": 29.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.09369004517793655,
-      "eval_runtime": 4.0266,
-      "eval_samples_per_second": 13.411,
-      "eval_steps_per_second": 0.497,
+      "eval_accuracy": 0.9074074074074074,
+      "eval_loss": 0.2718075215816498,
+      "eval_runtime": 4.0783,
+      "eval_samples_per_second": 13.241,
+      "eval_steps_per_second": 0.49,
       "step": 90
     },
     {
       "epoch": 30.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.09307903051376343,
-      "eval_runtime": 3.8076,
-      "eval_samples_per_second": 14.182,
-      "eval_steps_per_second": 0.525,
+      "eval_accuracy": 0.9074074074074074,
+      "eval_loss": 0.2738485634326935,
+      "eval_runtime": 3.7309,
+      "eval_samples_per_second": 14.474,
+      "eval_steps_per_second": 0.536,
       "step": 93
     },
     {
       "epoch": 31.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.09316574782133102,
-      "eval_runtime": 3.9741,
-      "eval_samples_per_second": 13.588,
-      "eval_steps_per_second": 0.503,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.26387789845466614,
+      "eval_runtime": 4.0527,
+      "eval_samples_per_second": 13.324,
+      "eval_steps_per_second": 0.493,
       "step": 96
     },
     {
       "epoch": 32.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.0940571129322052,
-      "eval_runtime": 4.072,
-      "eval_samples_per_second": 13.261,
-      "eval_steps_per_second": 0.491,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.25608906149864197,
+      "eval_runtime": 4.0393,
+      "eval_samples_per_second": 13.369,
+      "eval_steps_per_second": 0.495,
       "step": 99
     },
     {
       "epoch": 33.27,
       "learning_rate": 9.259259259259259e-06,
-      "loss": 0.3239,
+      "loss": 0.3407,
       "step": 100
     },
     {
       "epoch": 33.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.09196292608976364,
-      "eval_runtime": 3.764,
-      "eval_samples_per_second": 14.347,
-      "eval_steps_per_second": 0.531,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.2497122436761856,
+      "eval_runtime": 3.7801,
+      "eval_samples_per_second": 14.285,
+      "eval_steps_per_second": 0.529,
       "step": 102
     },
     {
       "epoch": 34.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.08508706837892532,
-      "eval_runtime": 3.9778,
-      "eval_samples_per_second": 13.575,
-      "eval_steps_per_second": 0.503,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.2501243054866791,
+      "eval_runtime": 4.0375,
+      "eval_samples_per_second": 13.374,
+      "eval_steps_per_second": 0.495,
       "step": 105
     },
     {
       "epoch": 35.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.08276030421257019,
-      "eval_runtime": 4.0417,
-      "eval_samples_per_second": 13.361,
-      "eval_steps_per_second": 0.495,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.24550123512744904,
+      "eval_runtime": 4.0543,
+      "eval_samples_per_second": 13.319,
+      "eval_steps_per_second": 0.493,
       "step": 108
     },
     {
       "epoch": 36.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.0810302197933197,
-      "eval_runtime": 3.7281,
-      "eval_samples_per_second": 14.485,
-      "eval_steps_per_second": 0.536,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.238052099943161,
+      "eval_runtime": 3.8053,
+      "eval_samples_per_second": 14.191,
+      "eval_steps_per_second": 0.526,
       "step": 111
     },
     {
       "epoch": 37.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.0800759419798851,
-      "eval_runtime": 4.1175,
-      "eval_samples_per_second": 13.115,
-      "eval_steps_per_second": 0.486,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.23396454751491547,
+      "eval_runtime": 4.0536,
+      "eval_samples_per_second": 13.321,
+      "eval_steps_per_second": 0.493,
       "step": 114
     },
     {
       "epoch": 38.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.08043589442968369,
-      "eval_runtime": 3.9928,
-      "eval_samples_per_second": 13.524,
-      "eval_steps_per_second": 0.501,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.23214255273342133,
+      "eval_runtime": 4.0625,
+      "eval_samples_per_second": 13.292,
+      "eval_steps_per_second": 0.492,
       "step": 117
     },
     {
       "epoch": 39.8,
       "learning_rate": 0.0,
-      "loss": 0.3111,
+      "loss": 0.3112,
       "step": 120
     },
     {
       "epoch": 39.8,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.08069680631160736,
-      "eval_runtime": 3.8487,
-      "eval_samples_per_second": 14.031,
-      "eval_steps_per_second": 0.52,
+      "eval_accuracy": 0.9259259259259259,
+      "eval_loss": 0.23147787153720856,
+      "eval_runtime": 3.7365,
+      "eval_samples_per_second": 14.452,
+      "eval_steps_per_second": 0.535,
       "step": 120
     },
     {
       "epoch": 39.8,
       "step": 120,
       "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
+      "train_loss": 0.6987951993942261,
+      "train_runtime": 2118.2023,
+      "train_samples_per_second": 9.045,
+      "train_steps_per_second": 0.057
     }
   ],
   "max_steps": 120,