End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 39.94,
-    "eval_accuracy": 0.9491525423728814,
-    "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
+    "epoch": 59.94,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.028991766273975372,
+    "eval_runtime": 3.8464,
+    "eval_samples_per_second": 15.339,
+    "eval_steps_per_second": 0.52,
+    "total_flos": 7.88616702456404e+17,
+    "train_loss": 0.4937874893347422,
+    "train_runtime": 3820.3512,
+    "train_samples_per_second": 8.308,
+    "train_steps_per_second": 0.063
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 39.94,
-    "eval_accuracy": 0.9491525423728814,
-    "eval_loss": 0.17896686494350433,
-    "eval_runtime": 4.0515,
-    "eval_samples_per_second": 14.562,
-    "eval_steps_per_second": 0.494
+    "epoch": 59.94,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.028991766273975372,
+    "eval_runtime": 3.8464,
+    "eval_samples_per_second": 15.339,
+    "eval_steps_per_second": 0.52
 }

runs/Dec21_12-30-01_93f77f96eef0/events.out.tfevents.1671629924.93f77f96eef0.928.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69d0d4da283bca6151dcccfde78f9d8488481ef05569af9660ea668856b02886
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "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
+    "epoch": 59.94,
+    "total_flos": 7.88616702456404e+17,
+    "train_loss": 0.4937874893347422,
+    "train_runtime": 3820.3512,
+    "train_samples_per_second": 8.308,
+    "train_steps_per_second": 0.063
 }

trainer_state.json

@@ -1,433 +1,637 @@
 {
-  "best_metric": 0.9491525423728814,
-  "best_model_checkpoint": "delivery_truck_classification/checkpoint-52",
-  "epoch": 39.94117647058823,
-  "global_step": 160,
+  "best_metric": 1.0,
+  "best_model_checkpoint": "delivery_truck_classification/checkpoint-172",
+  "epoch": 59.94117647058823,
+  "global_step": 240,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "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,
+      "eval_accuracy": 0.1864406779661017,
+      "eval_loss": 1.9123966693878174,
+      "eval_runtime": 3.7131,
+      "eval_samples_per_second": 15.89,
+      "eval_steps_per_second": 0.539,
       "step": 4
     },
     {
       "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,
+      "eval_accuracy": 0.23728813559322035,
+      "eval_loss": 1.8094592094421387,
+      "eval_runtime": 4.4625,
+      "eval_samples_per_second": 13.221,
+      "eval_steps_per_second": 0.448,
       "step": 8
     },
     {
       "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,
+      "eval_accuracy": 0.3898305084745763,
+      "eval_loss": 1.6757333278656006,
+      "eval_runtime": 3.88,
+      "eval_samples_per_second": 15.206,
+      "eval_steps_per_second": 0.515,
       "step": 12
     },
     {
       "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,
+      "eval_accuracy": 0.5254237288135594,
+      "eval_loss": 1.4906411170959473,
+      "eval_runtime": 3.788,
+      "eval_samples_per_second": 15.576,
+      "eval_steps_per_second": 0.528,
       "step": 16
     },
     {
       "epoch": 4.94,
-      "learning_rate": 4.8611111111111115e-05,
-      "loss": 1.7242,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 1.8286,
       "step": 20
     },
     {
       "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,
+      "eval_accuracy": 0.6440677966101694,
+      "eval_loss": 1.2703886032104492,
+      "eval_runtime": 4.208,
+      "eval_samples_per_second": 14.021,
+      "eval_steps_per_second": 0.475,
       "step": 20
     },
     {
       "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,
+      "eval_accuracy": 0.6779661016949152,
+      "eval_loss": 1.0685124397277832,
+      "eval_runtime": 3.8096,
+      "eval_samples_per_second": 15.487,
+      "eval_steps_per_second": 0.525,
       "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,
+      "eval_accuracy": 0.7457627118644068,
+      "eval_loss": 0.8031855225563049,
+      "eval_runtime": 4.1542,
+      "eval_samples_per_second": 14.203,
+      "eval_steps_per_second": 0.481,
       "step": 28
     },
     {
       "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,
+      "eval_accuracy": 0.7627118644067796,
+      "eval_loss": 0.6308717131614685,
+      "eval_runtime": 3.7602,
+      "eval_samples_per_second": 15.691,
+      "eval_steps_per_second": 0.532,
       "step": 32
     },
     {
       "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,
+      "eval_accuracy": 0.847457627118644,
+      "eval_loss": 0.49890831112861633,
+      "eval_runtime": 4.1569,
+      "eval_samples_per_second": 14.193,
+      "eval_steps_per_second": 0.481,
       "step": 36
     },
     {
       "epoch": 9.94,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 0.8337,
+      "learning_rate": 4.62962962962963e-05,
+      "loss": 0.9342,
       "step": 40
     },
     {
       "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,
+      "eval_accuracy": 0.847457627118644,
+      "eval_loss": 0.40634387731552124,
+      "eval_runtime": 3.8202,
+      "eval_samples_per_second": 15.444,
+      "eval_steps_per_second": 0.524,
       "step": 40
     },
     {
       "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,
+      "eval_accuracy": 0.9152542372881356,
+      "eval_loss": 0.26924625039100647,
+      "eval_runtime": 4.1384,
+      "eval_samples_per_second": 14.257,
+      "eval_steps_per_second": 0.483,
       "step": 44
     },
     {
       "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,
+      "eval_accuracy": 0.8983050847457628,
+      "eval_loss": 0.27362605929374695,
+      "eval_runtime": 3.707,
+      "eval_samples_per_second": 15.916,
+      "eval_steps_per_second": 0.54,
       "step": 48
     },
     {
       "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,
+      "eval_accuracy": 0.9322033898305084,
+      "eval_loss": 0.21159349381923676,
+      "eval_runtime": 4.0712,
+      "eval_samples_per_second": 14.492,
+      "eval_steps_per_second": 0.491,
       "step": 52
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.1497751921415329,
+      "eval_runtime": 3.629,
+      "eval_samples_per_second": 16.258,
+      "eval_steps_per_second": 0.551,
       "step": 56
     },
     {
       "epoch": 14.94,
-      "learning_rate": 3.472222222222222e-05,
-      "loss": 0.493,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.5151,
       "step": 60
     },
     {
       "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,
+      "eval_loss": 0.1906394362449646,
+      "eval_runtime": 4.159,
+      "eval_samples_per_second": 14.186,
+      "eval_steps_per_second": 0.481,
       "step": 60
     },
     {
       "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,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.16975143551826477,
+      "eval_runtime": 3.7313,
+      "eval_samples_per_second": 15.812,
+      "eval_steps_per_second": 0.536,
       "step": 64
     },
     {
       "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,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.1431875377893448,
+      "eval_runtime": 4.1272,
+      "eval_samples_per_second": 14.295,
+      "eval_steps_per_second": 0.485,
       "step": 68
     },
     {
       "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,
+      "eval_loss": 0.16815055906772614,
+      "eval_runtime": 3.6724,
+      "eval_samples_per_second": 16.066,
+      "eval_steps_per_second": 0.545,
       "step": 72
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.10692509263753891,
+      "eval_runtime": 4.1232,
+      "eval_samples_per_second": 14.309,
+      "eval_steps_per_second": 0.485,
       "step": 76
     },
     {
       "epoch": 19.94,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.3966,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.4009,
       "step": 80
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.08214119076728821,
+      "eval_runtime": 3.6886,
+      "eval_samples_per_second": 15.995,
+      "eval_steps_per_second": 0.542,
       "step": 80
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.09025811403989792,
+      "eval_runtime": 4.3053,
+      "eval_samples_per_second": 13.704,
+      "eval_steps_per_second": 0.465,
       "step": 84
     },
     {
       "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,
+      "eval_accuracy": 0.9661016949152542,
+      "eval_loss": 0.12810270488262177,
+      "eval_runtime": 3.7203,
+      "eval_samples_per_second": 15.859,
+      "eval_steps_per_second": 0.538,
       "step": 88
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.09356144815683365,
+      "eval_runtime": 4.118,
+      "eval_samples_per_second": 14.328,
+      "eval_steps_per_second": 0.486,
       "step": 92
     },
     {
       "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,
+      "eval_accuracy": 0.9661016949152542,
+      "eval_loss": 0.10591834783554077,
+      "eval_runtime": 3.6933,
+      "eval_samples_per_second": 15.975,
+      "eval_steps_per_second": 0.542,
       "step": 96
     },
     {
       "epoch": 24.94,
-      "learning_rate": 2.0833333333333336e-05,
-      "loss": 0.3645,
+      "learning_rate": 3.240740740740741e-05,
+      "loss": 0.3482,
       "step": 100
     },
     {
       "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,
+      "eval_accuracy": 0.9491525423728814,
+      "eval_loss": 0.14312368631362915,
+      "eval_runtime": 4.1639,
+      "eval_samples_per_second": 14.169,
+      "eval_steps_per_second": 0.48,
       "step": 100
     },
     {
       "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,
+      "eval_accuracy": 0.9661016949152542,
+      "eval_loss": 0.08986904472112656,
+      "eval_runtime": 3.7088,
+      "eval_samples_per_second": 15.908,
+      "eval_steps_per_second": 0.539,
       "step": 104
     },
     {
       "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,
+      "eval_accuracy": 0.9661016949152542,
+      "eval_loss": 0.06888163089752197,
+      "eval_runtime": 4.1304,
+      "eval_samples_per_second": 14.284,
+      "eval_steps_per_second": 0.484,
       "step": 108
     },
     {
       "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,
+      "eval_accuracy": 0.9661016949152542,
+      "eval_loss": 0.07510155439376831,
+      "eval_runtime": 3.7462,
+      "eval_samples_per_second": 15.749,
+      "eval_steps_per_second": 0.534,
       "step": 112
     },
     {
       "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,
+      "eval_accuracy": 0.9661016949152542,
+      "eval_loss": 0.08912018686532974,
+      "eval_runtime": 3.7992,
+      "eval_samples_per_second": 15.53,
+      "eval_steps_per_second": 0.526,
       "step": 116
     },
     {
       "epoch": 29.94,
-      "learning_rate": 1.388888888888889e-05,
-      "loss": 0.3358,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.3306,
       "step": 120
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.05226200073957443,
+      "eval_runtime": 4.2109,
+      "eval_samples_per_second": 14.011,
+      "eval_steps_per_second": 0.475,
       "step": 120
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.07344209402799606,
+      "eval_runtime": 3.9025,
+      "eval_samples_per_second": 15.118,
+      "eval_steps_per_second": 0.512,
       "step": 124
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.07462318241596222,
+      "eval_runtime": 4.1592,
+      "eval_samples_per_second": 14.185,
+      "eval_steps_per_second": 0.481,
       "step": 128
     },
     {
       "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,
+      "eval_accuracy": 0.9661016949152542,
+      "eval_loss": 0.04742526262998581,
+      "eval_runtime": 3.8091,
+      "eval_samples_per_second": 15.489,
+      "eval_steps_per_second": 0.525,
       "step": 132
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.044318120926618576,
+      "eval_runtime": 3.7853,
+      "eval_samples_per_second": 15.587,
+      "eval_steps_per_second": 0.528,
       "step": 136
     },
     {
       "epoch": 34.94,
-      "learning_rate": 6.944444444444445e-06,
-      "loss": 0.3049,
+      "learning_rate": 2.314814814814815e-05,
+      "loss": 0.2871,
       "step": 140
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.08141448348760605,
+      "eval_runtime": 4.1875,
+      "eval_samples_per_second": 14.09,
+      "eval_steps_per_second": 0.478,
       "step": 140
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.0690697729587555,
+      "eval_runtime": 3.7869,
+      "eval_samples_per_second": 15.58,
+      "eval_steps_per_second": 0.528,
       "step": 144
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.053145911544561386,
+      "eval_runtime": 4.1845,
+      "eval_samples_per_second": 14.1,
+      "eval_steps_per_second": 0.478,
       "step": 148
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.06143573671579361,
+      "eval_runtime": 3.7811,
+      "eval_samples_per_second": 15.604,
+      "eval_steps_per_second": 0.529,
       "step": 152
     },
     {
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.057755034416913986,
+      "eval_runtime": 4.2237,
+      "eval_samples_per_second": 13.969,
+      "eval_steps_per_second": 0.474,
       "step": 156
     },
     {
       "epoch": 39.94,
-      "learning_rate": 0.0,
-      "loss": 0.2962,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.2754,
       "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,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.05201863497495651,
+      "eval_runtime": 3.7528,
+      "eval_samples_per_second": 15.722,
+      "eval_steps_per_second": 0.533,
       "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
+      "epoch": 40.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.05365251377224922,
+      "eval_runtime": 3.7906,
+      "eval_samples_per_second": 15.565,
+      "eval_steps_per_second": 0.528,
+      "step": 164
+    },
+    {
+      "epoch": 41.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.04466838762164116,
+      "eval_runtime": 4.0883,
+      "eval_samples_per_second": 14.432,
+      "eval_steps_per_second": 0.489,
+      "step": 168
+    },
+    {
+      "epoch": 42.94,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.028991766273975372,
+      "eval_runtime": 3.7894,
+      "eval_samples_per_second": 15.57,
+      "eval_steps_per_second": 0.528,
+      "step": 172
+    },
+    {
+      "epoch": 43.94,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.029105449095368385,
+      "eval_runtime": 4.1829,
+      "eval_samples_per_second": 14.105,
+      "eval_steps_per_second": 0.478,
+      "step": 176
+    },
+    {
+      "epoch": 44.94,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.269,
+      "step": 180
+    },
+    {
+      "epoch": 44.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03256472200155258,
+      "eval_runtime": 3.7585,
+      "eval_samples_per_second": 15.698,
+      "eval_steps_per_second": 0.532,
+      "step": 180
+    },
+    {
+      "epoch": 45.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03297417610883713,
+      "eval_runtime": 4.155,
+      "eval_samples_per_second": 14.2,
+      "eval_steps_per_second": 0.481,
+      "step": 184
+    },
+    {
+      "epoch": 46.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.034774668514728546,
+      "eval_runtime": 3.7675,
+      "eval_samples_per_second": 15.66,
+      "eval_steps_per_second": 0.531,
+      "step": 188
+    },
+    {
+      "epoch": 47.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03467889130115509,
+      "eval_runtime": 4.0766,
+      "eval_samples_per_second": 14.473,
+      "eval_steps_per_second": 0.491,
+      "step": 192
+    },
+    {
+      "epoch": 48.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03474270924925804,
+      "eval_runtime": 3.7643,
+      "eval_samples_per_second": 15.674,
+      "eval_steps_per_second": 0.531,
+      "step": 196
+    },
+    {
+      "epoch": 49.94,
+      "learning_rate": 9.259259259259259e-06,
+      "loss": 0.2615,
+      "step": 200
+    },
+    {
+      "epoch": 49.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.04243284463882446,
+      "eval_runtime": 4.0985,
+      "eval_samples_per_second": 14.395,
+      "eval_steps_per_second": 0.488,
+      "step": 200
+    },
+    {
+      "epoch": 50.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.04514499008655548,
+      "eval_runtime": 3.9491,
+      "eval_samples_per_second": 14.94,
+      "eval_steps_per_second": 0.506,
+      "step": 204
+    },
+    {
+      "epoch": 51.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.043262287974357605,
+      "eval_runtime": 3.6957,
+      "eval_samples_per_second": 15.964,
+      "eval_steps_per_second": 0.541,
+      "step": 208
+    },
+    {
+      "epoch": 52.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.035157691687345505,
+      "eval_runtime": 4.172,
+      "eval_samples_per_second": 14.142,
+      "eval_steps_per_second": 0.479,
+      "step": 212
+    },
+    {
+      "epoch": 53.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03394060581922531,
+      "eval_runtime": 3.7213,
+      "eval_samples_per_second": 15.855,
+      "eval_steps_per_second": 0.537,
+      "step": 216
+    },
+    {
+      "epoch": 54.94,
+      "learning_rate": 4.6296296296296296e-06,
+      "loss": 0.2386,
+      "step": 220
+    },
+    {
+      "epoch": 54.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03394594043493271,
+      "eval_runtime": 4.1524,
+      "eval_samples_per_second": 14.208,
+      "eval_steps_per_second": 0.482,
+      "step": 220
+    },
+    {
+      "epoch": 55.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03388730436563492,
+      "eval_runtime": 3.719,
+      "eval_samples_per_second": 15.864,
+      "eval_steps_per_second": 0.538,
+      "step": 224
+    },
+    {
+      "epoch": 56.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.034843575209379196,
+      "eval_runtime": 4.1514,
+      "eval_samples_per_second": 14.212,
+      "eval_steps_per_second": 0.482,
+      "step": 228
+    },
+    {
+      "epoch": 57.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03663792461156845,
+      "eval_runtime": 3.7689,
+      "eval_samples_per_second": 15.654,
+      "eval_steps_per_second": 0.531,
+      "step": 232
+    },
+    {
+      "epoch": 58.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03743705153465271,
+      "eval_runtime": 4.1165,
+      "eval_samples_per_second": 14.333,
+      "eval_steps_per_second": 0.486,
+      "step": 236
+    },
+    {
+      "epoch": 59.94,
+      "learning_rate": 0.0,
+      "loss": 0.2362,
+      "step": 240
+    },
+    {
+      "epoch": 59.94,
+      "eval_accuracy": 0.9830508474576272,
+      "eval_loss": 0.03746996819972992,
+      "eval_runtime": 3.7404,
+      "eval_samples_per_second": 15.774,
+      "eval_steps_per_second": 0.535,
+      "step": 240
+    },
+    {
+      "epoch": 59.94,
+      "step": 240,
+      "total_flos": 7.88616702456404e+17,
+      "train_loss": 0.4937874893347422,
+      "train_runtime": 3820.3512,
+      "train_samples_per_second": 8.308,
+      "train_steps_per_second": 0.063
     }
   ],
-  "max_steps": 160,
-  "num_train_epochs": 40,
-  "total_flos": 5.226950410064732e+17,
+  "max_steps": 240,
+  "num_train_epochs": 60,
+  "total_flos": 7.88616702456404e+17,
   "trial_name": null,
   "trial_params": null
 }