End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 39.8,
+    "epoch": 39.94,
     "eval_accuracy": 0.9491525423728814,
-    "eval_loss": 0.1655263751745224,
-    "eval_runtime": 3.8793,
-    "eval_samples_per_second": 13.92,
-    "eval_steps_per_second": 0.516,
-    "total_flos": 4.7393239534534656e+17,
-    "train_loss": 0.6987951993942261,
-    "train_runtime": 2118.2023,
-    "train_samples_per_second": 9.045,
-    "train_steps_per_second": 0.057
+    "eval_loss": 0.17896686494350433,
+    "eval_runtime": 4.0515,
+    "eval_samples_per_second": 14.562,
+    "eval_steps_per_second": 0.494,
+    "total_flos": 5.226950410064732e+17,
+    "train_loss": 0.5936037123203277,
+    "train_runtime": 2238.3601,
+    "train_samples_per_second": 9.4,
+    "train_steps_per_second": 0.071
 }

eval_results.json

@@ -1,4 +1,8 @@
 {
+    "epoch": 39.94,
     "eval_accuracy": 0.9491525423728814,
-    "eval_loss": 0.1655263751745224
+    "eval_loss": 0.17896686494350433,
+    "eval_runtime": 4.0515,
+    "eval_samples_per_second": 14.562,
+    "eval_steps_per_second": 0.494
 }

runs/Dec20_20-35-32_ece0298277ec/events.out.tfevents.1671571028.ece0298277ec.230.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:febcd5bfb310f75b78c8d586dbf520b7609090062c057fa865586fbeace09aad
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 39.8,
-    "total_flos": 4.7393239534534656e+17,
-    "train_loss": 0.6987951993942261,
-    "train_runtime": 2118.2023,
-    "train_samples_per_second": 9.045,
-    "train_steps_per_second": 0.057
+    "epoch": 39.94,
+    "total_flos": 5.226950410064732e+17,
+    "train_loss": 0.5936037123203277,
+    "train_runtime": 2238.3601,
+    "train_samples_per_second": 9.4,
+    "train_steps_per_second": 0.071
 }

trainer_state.json

@@ -1,421 +1,433 @@
 {
-  "best_metric": 0.9259259259259259,
-  "best_model_checkpoint": "delivery_truck_classification/checkpoint-66",
-  "epoch": 39.8,
-  "global_step": 120,
+  "best_metric": 0.9491525423728814,
+  "best_model_checkpoint": "delivery_truck_classification/checkpoint-52",
+  "epoch": 39.94117647058823,
+  "global_step": 160,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.8,
-      "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": 0.25925925925925924,
-      "eval_loss": 1.7420549392700195,
-      "eval_runtime": 4.3749,
-      "eval_samples_per_second": 12.343,
-      "eval_steps_per_second": 0.457,
-      "step": 6
+      "epoch": 0.94,
+      "eval_accuracy": 0.11864406779661017,
+      "eval_loss": 1.888161540031433,
+      "eval_runtime": 4.1522,
+      "eval_samples_per_second": 14.209,
+      "eval_steps_per_second": 0.482,
+      "step": 4
     },
     {
-      "epoch": 2.8,
-      "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": 1.94,
+      "eval_accuracy": 0.3559322033898305,
+      "eval_loss": 1.67988920211792,
+      "eval_runtime": 4.5824,
+      "eval_samples_per_second": 12.875,
+      "eval_steps_per_second": 0.436,
+      "step": 8
     },
     {
-      "epoch": 3.8,
-      "eval_accuracy": 0.5,
-      "eval_loss": 1.4371126890182495,
-      "eval_runtime": 3.8435,
-      "eval_samples_per_second": 14.05,
-      "eval_steps_per_second": 0.52,
+      "epoch": 2.94,
+      "eval_accuracy": 0.576271186440678,
+      "eval_loss": 1.4259672164916992,
+      "eval_runtime": 3.9572,
+      "eval_samples_per_second": 14.91,
+      "eval_steps_per_second": 0.505,
       "step": 12
     },
     {
-      "epoch": 4.8,
-      "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": 3.94,
+      "eval_accuracy": 0.6779661016949152,
+      "eval_loss": 1.1092320680618286,
+      "eval_runtime": 4.365,
+      "eval_samples_per_second": 13.517,
+      "eval_steps_per_second": 0.458,
+      "step": 16
     },
     {
-      "epoch": 5.8,
-      "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": 4.94,
+      "learning_rate": 4.8611111111111115e-05,
+      "loss": 1.7242,
+      "step": 20
     },
     {
-      "epoch": 6.53,
-      "learning_rate": 4.62962962962963e-05,
-      "loss": 1.8164,
+      "epoch": 4.94,
+      "eval_accuracy": 0.7457627118644068,
+      "eval_loss": 0.8652871251106262,
+      "eval_runtime": 3.9966,
+      "eval_samples_per_second": 14.762,
+      "eval_steps_per_second": 0.5,
       "step": 20
     },
     {
-      "epoch": 6.8,
-      "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": 5.94,
+      "eval_accuracy": 0.7796610169491526,
+      "eval_loss": 0.6787465810775757,
+      "eval_runtime": 4.26,
+      "eval_samples_per_second": 13.85,
+      "eval_steps_per_second": 0.469,
+      "step": 24
     },
     {
-      "epoch": 7.8,
-      "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": 6.94,
+      "eval_accuracy": 0.8305084745762712,
+      "eval_loss": 0.5506174564361572,
+      "eval_runtime": 4.3376,
+      "eval_samples_per_second": 13.602,
+      "eval_steps_per_second": 0.461,
+      "step": 28
     },
     {
-      "epoch": 8.8,
-      "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": 7.94,
+      "eval_accuracy": 0.8813559322033898,
+      "eval_loss": 0.4173695743083954,
+      "eval_runtime": 3.9498,
+      "eval_samples_per_second": 14.937,
+      "eval_steps_per_second": 0.506,
+      "step": 32
     },
     {
-      "epoch": 9.8,
-      "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": 8.94,
+      "eval_accuracy": 0.8813559322033898,
+      "eval_loss": 0.3643472194671631,
+      "eval_runtime": 4.2828,
+      "eval_samples_per_second": 13.776,
+      "eval_steps_per_second": 0.467,
+      "step": 36
     },
     {
-      "epoch": 10.8,
-      "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": 9.94,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.8337,
+      "step": 40
     },
     {
-      "epoch": 11.8,
-      "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": 9.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.2679864466190338,
+      "eval_runtime": 4.0645,
+      "eval_samples_per_second": 14.516,
+      "eval_steps_per_second": 0.492,
+      "step": 40
     },
     {
-      "epoch": 12.8,
-      "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": 10.94,
+      "eval_accuracy": 0.8983050847457628,
+      "eval_loss": 0.2705465257167816,
+      "eval_runtime": 3.9127,
+      "eval_samples_per_second": 15.079,
+      "eval_steps_per_second": 0.511,
+      "step": 44
     },
     {
-      "epoch": 13.27,
-      "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.8525,
-      "step": 40
+      "epoch": 11.94,
+      "eval_accuracy": 0.9152542372881356,
+      "eval_loss": 0.2270025759935379,
+      "eval_runtime": 4.2904,
+      "eval_samples_per_second": 13.752,
+      "eval_steps_per_second": 0.466,
+      "step": 48
     },
     {
-      "epoch": 13.8,
-      "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": 12.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.17896686494350433,
+      "eval_runtime": 4.2572,
+      "eval_samples_per_second": 13.859,
+      "eval_steps_per_second": 0.47,
+      "step": 52
     },
     {
-      "epoch": 14.8,
-      "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": 13.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.16937005519866943,
+      "eval_runtime": 3.9569,
+      "eval_samples_per_second": 14.911,
+      "eval_steps_per_second": 0.505,
+      "step": 56
     },
     {
-      "epoch": 15.8,
-      "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": 14.94,
+      "learning_rate": 3.472222222222222e-05,
+      "loss": 0.493,
+      "step": 60
     },
     {
-      "epoch": 16.8,
-      "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": 14.94,
+      "eval_accuracy": 0.9152542372881356,
+      "eval_loss": 0.17763535678386688,
+      "eval_runtime": 4.2916,
+      "eval_samples_per_second": 13.748,
+      "eval_steps_per_second": 0.466,
+      "step": 60
     },
     {
-      "epoch": 17.8,
-      "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": 15.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.1831100881099701,
+      "eval_runtime": 4.3067,
+      "eval_samples_per_second": 13.7,
+      "eval_steps_per_second": 0.464,
+      "step": 64
     },
     {
-      "epoch": 18.8,
-      "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": 16.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.17647486925125122,
+      "eval_runtime": 3.9721,
+      "eval_samples_per_second": 14.854,
+      "eval_steps_per_second": 0.504,
+      "step": 68
     },
     {
-      "epoch": 19.8,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.4831,
-      "step": 60
+      "epoch": 17.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.15746894478797913,
+      "eval_runtime": 4.3018,
+      "eval_samples_per_second": 13.715,
+      "eval_steps_per_second": 0.465,
+      "step": 72
     },
     {
-      "epoch": 19.8,
-      "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": 18.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.14716410636901855,
+      "eval_runtime": 4.2957,
+      "eval_samples_per_second": 13.735,
+      "eval_steps_per_second": 0.466,
+      "step": 76
+    },
+    {
+      "epoch": 19.94,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.3966,
+      "step": 80
     },
     {
-      "epoch": 20.8,
-      "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": 19.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.13601559400558472,
+      "eval_runtime": 3.8795,
+      "eval_samples_per_second": 15.208,
+      "eval_steps_per_second": 0.516,
+      "step": 80
     },
     {
-      "epoch": 21.8,
-      "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": 20.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.1448136419057846,
+      "eval_runtime": 4.2669,
+      "eval_samples_per_second": 13.827,
+      "eval_steps_per_second": 0.469,
+      "step": 84
     },
     {
-      "epoch": 22.8,
-      "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": 21.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.16579066216945648,
+      "eval_runtime": 4.1904,
+      "eval_samples_per_second": 14.08,
+      "eval_steps_per_second": 0.477,
+      "step": 88
     },
     {
-      "epoch": 23.8,
-      "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": 22.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.1652187556028366,
+      "eval_runtime": 3.894,
+      "eval_samples_per_second": 15.152,
+      "eval_steps_per_second": 0.514,
+      "step": 92
     },
     {
-      "epoch": 24.8,
-      "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": 23.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.15646469593048096,
+      "eval_runtime": 4.3539,
+      "eval_samples_per_second": 13.551,
+      "eval_steps_per_second": 0.459,
+      "step": 96
     },
     {
-      "epoch": 25.8,
-      "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": 24.94,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 0.3645,
+      "step": 100
     },
     {
-      "epoch": 26.53,
-      "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.3889,
-      "step": 80
+      "epoch": 24.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.1700868457555771,
+      "eval_runtime": 3.9704,
+      "eval_samples_per_second": 14.86,
+      "eval_steps_per_second": 0.504,
+      "step": 100
     },
     {
-      "epoch": 26.8,
-      "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": 25.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.18298125267028809,
+      "eval_runtime": 3.9238,
+      "eval_samples_per_second": 15.036,
+      "eval_steps_per_second": 0.51,
+      "step": 104
     },
     {
-      "epoch": 27.8,
-      "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": 26.94,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.16823288798332214,
+      "eval_runtime": 5.0017,
+      "eval_samples_per_second": 11.796,
+      "eval_steps_per_second": 0.4,
+      "step": 108
     },
     {
-      "epoch": 28.8,
-      "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": 27.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.1410430371761322,
+      "eval_runtime": 4.472,
+      "eval_samples_per_second": 13.193,
+      "eval_steps_per_second": 0.447,
+      "step": 112
     },
     {
-      "epoch": 29.8,
-      "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": 28.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12907052040100098,
+      "eval_runtime": 3.8982,
+      "eval_samples_per_second": 15.135,
+      "eval_steps_per_second": 0.513,
+      "step": 116
     },
     {
-      "epoch": 30.8,
-      "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": 29.94,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.3358,
+      "step": 120
     },
     {
-      "epoch": 31.8,
-      "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": 29.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12478938698768616,
+      "eval_runtime": 4.248,
+      "eval_samples_per_second": 13.889,
+      "eval_steps_per_second": 0.471,
+      "step": 120
     },
     {
-      "epoch": 32.8,
-      "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": 30.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12751266360282898,
+      "eval_runtime": 4.3163,
+      "eval_samples_per_second": 13.669,
+      "eval_steps_per_second": 0.463,
+      "step": 124
     },
     {
-      "epoch": 33.27,
-      "learning_rate": 9.259259259259259e-06,
-      "loss": 0.3407,
-      "step": 100
+      "epoch": 31.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.1257278025150299,
+      "eval_runtime": 3.9781,
+      "eval_samples_per_second": 14.831,
+      "eval_steps_per_second": 0.503,
+      "step": 128
     },
     {
-      "epoch": 33.8,
-      "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": 32.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12883280217647552,
+      "eval_runtime": 4.3885,
+      "eval_samples_per_second": 13.444,
+      "eval_steps_per_second": 0.456,
+      "step": 132
     },
     {
-      "epoch": 34.8,
-      "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": 33.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12463054060935974,
+      "eval_runtime": 4.1226,
+      "eval_samples_per_second": 14.311,
+      "eval_steps_per_second": 0.485,
+      "step": 136
     },
     {
-      "epoch": 35.8,
-      "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": 34.94,
+      "learning_rate": 6.944444444444445e-06,
+      "loss": 0.3049,
+      "step": 140
     },
     {
-      "epoch": 36.8,
-      "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": 34.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12190988659858704,
+      "eval_runtime": 3.969,
+      "eval_samples_per_second": 14.865,
+      "eval_steps_per_second": 0.504,
+      "step": 140
     },
     {
-      "epoch": 37.8,
-      "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": 35.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12239754945039749,
+      "eval_runtime": 4.3753,
+      "eval_samples_per_second": 13.485,
+      "eval_steps_per_second": 0.457,
+      "step": 144
     },
     {
-      "epoch": 38.8,
-      "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": 36.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12457720935344696,
+      "eval_runtime": 4.1325,
+      "eval_samples_per_second": 14.277,
+      "eval_steps_per_second": 0.484,
+      "step": 148
     },
     {
-      "epoch": 39.8,
-      "learning_rate": 0.0,
-      "loss": 0.3112,
-      "step": 120
+      "epoch": 37.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12425000220537186,
+      "eval_runtime": 3.9849,
+      "eval_samples_per_second": 14.806,
+      "eval_steps_per_second": 0.502,
+      "step": 152
     },
     {
-      "epoch": 39.8,
-      "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": 38.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12478892505168915,
+      "eval_runtime": 4.257,
+      "eval_samples_per_second": 13.86,
+      "eval_steps_per_second": 0.47,
+      "step": 156
     },
     {
-      "epoch": 39.8,
-      "step": 120,
-      "total_flos": 4.7393239534534656e+17,
-      "train_loss": 0.6987951993942261,
-      "train_runtime": 2118.2023,
-      "train_samples_per_second": 9.045,
-      "train_steps_per_second": 0.057
+      "epoch": 39.94,
+      "learning_rate": 0.0,
+      "loss": 0.2962,
+      "step": 160
+    },
+    {
+      "epoch": 39.94,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.12530554831027985,
+      "eval_runtime": 4.3582,
+      "eval_samples_per_second": 13.538,
+      "eval_steps_per_second": 0.459,
+      "step": 160
+    },
+    {
+      "epoch": 39.94,
+      "step": 160,
+      "total_flos": 5.226950410064732e+17,
+      "train_loss": 0.5936037123203277,
+      "train_runtime": 2238.3601,
+      "train_samples_per_second": 9.4,
+      "train_steps_per_second": 0.071
     }
   ],
-  "max_steps": 120,
+  "max_steps": 160,
   "num_train_epochs": 40,
-  "total_flos": 4.7393239534534656e+17,
+  "total_flos": 5.226950410064732e+17,
   "trial_name": null,
   "trial_params": null
 }