End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 39.92,
-    "eval_accuracy": 0.9555555555555556,
-    "eval_loss": 0.20596002042293549,
-    "eval_runtime": 3.0656,
-    "eval_samples_per_second": 14.679,
-    "eval_steps_per_second": 0.652,
-    "total_flos": 3.992815176520827e+17,
-    "train_loss": 0.1958512246608734,
-    "train_runtime": 1830.7134,
-    "train_samples_per_second": 8.783,
-    "train_steps_per_second": 0.066
+    "epoch": 39.86,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.062098320573568344,
+    "eval_runtime": 3.1556,
+    "eval_samples_per_second": 14.894,
+    "eval_steps_per_second": 0.634,
+    "total_flos": 4.138238964234977e+17,
+    "train_loss": 0.19912103215853375,
+    "train_runtime": 1848.5966,
+    "train_samples_per_second": 9.023,
+    "train_steps_per_second": 0.065
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 39.92,
-    "eval_accuracy": 0.9555555555555556,
-    "eval_loss": 0.20596002042293549,
-    "eval_runtime": 3.0656,
-    "eval_samples_per_second": 14.679,
-    "eval_steps_per_second": 0.652
+    "epoch": 39.86,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.062098320573568344,
+    "eval_runtime": 3.1556,
+    "eval_samples_per_second": 14.894,
+    "eval_steps_per_second": 0.634
 }

runs/Oct13_10-52-01_7f13ffb5462b/events.out.tfevents.1665660552.7f13ffb5462b.68.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a40fe4a7ed12135be3b0b4cf530b68485c06a027c2b63c27ea64ac223c9fb529
+size 357

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 39.92,
-    "total_flos": 3.992815176520827e+17,
-    "train_loss": 0.1958512246608734,
-    "train_runtime": 1830.7134,
-    "train_samples_per_second": 8.783,
-    "train_steps_per_second": 0.066
+    "epoch": 39.86,
+    "total_flos": 4.138238964234977e+17,
+    "train_loss": 0.19912103215853375,
+    "train_runtime": 1848.5966,
+    "train_samples_per_second": 9.023,
+    "train_steps_per_second": 0.065
 }

trainer_state.json

@@ -1,421 +1,421 @@
 {
-  "best_metric": 0.9555555555555556,
-  "best_model_checkpoint": "delivery_truck_classification/checkpoint-27",
-  "epoch": 39.92307692307692,
+  "best_metric": 1.0,
+  "best_model_checkpoint": "delivery_truck_classification/checkpoint-9",
+  "epoch": 39.857142857142854,
   "global_step": 120,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.92,
-      "eval_accuracy": 0.9111111111111111,
-      "eval_loss": 0.3269350230693817,
-      "eval_runtime": 3.0493,
-      "eval_samples_per_second": 14.757,
-      "eval_steps_per_second": 0.656,
+      "epoch": 0.86,
+      "eval_accuracy": 0.9574468085106383,
+      "eval_loss": 0.14558981359004974,
+      "eval_runtime": 3.0797,
+      "eval_samples_per_second": 15.261,
+      "eval_steps_per_second": 0.649,
       "step": 3
     },
     {
-      "epoch": 1.92,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.28137677907943726,
-      "eval_runtime": 3.5544,
-      "eval_samples_per_second": 12.66,
-      "eval_steps_per_second": 0.563,
+      "epoch": 1.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.08996303379535675,
+      "eval_runtime": 3.5219,
+      "eval_samples_per_second": 13.345,
+      "eval_steps_per_second": 0.568,
       "step": 6
     },
     {
-      "epoch": 2.92,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.26246681809425354,
-      "eval_runtime": 3.3113,
-      "eval_samples_per_second": 13.59,
-      "eval_steps_per_second": 0.604,
+      "epoch": 2.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.062098320573568344,
+      "eval_runtime": 3.3535,
+      "eval_samples_per_second": 14.015,
+      "eval_steps_per_second": 0.596,
       "step": 9
     },
     {
-      "epoch": 3.92,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.2771083414554596,
-      "eval_runtime": 2.9987,
-      "eval_samples_per_second": 15.006,
-      "eval_steps_per_second": 0.667,
+      "epoch": 3.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.06347361207008362,
+      "eval_runtime": 3.0769,
+      "eval_samples_per_second": 15.275,
+      "eval_steps_per_second": 0.65,
       "step": 12
     },
     {
-      "epoch": 4.92,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.24194401502609253,
-      "eval_runtime": 3.2862,
-      "eval_samples_per_second": 13.694,
-      "eval_steps_per_second": 0.609,
+      "epoch": 4.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.05354984849691391,
+      "eval_runtime": 3.0814,
+      "eval_samples_per_second": 15.253,
+      "eval_steps_per_second": 0.649,
       "step": 15
     },
     {
-      "epoch": 5.92,
-      "eval_accuracy": 0.9111111111111111,
-      "eval_loss": 0.2263958752155304,
-      "eval_runtime": 3.8859,
-      "eval_samples_per_second": 11.58,
-      "eval_steps_per_second": 0.515,
+      "epoch": 5.86,
+      "eval_accuracy": 0.9574468085106383,
+      "eval_loss": 0.06630665063858032,
+      "eval_runtime": 3.3475,
+      "eval_samples_per_second": 14.04,
+      "eval_steps_per_second": 0.597,
       "step": 18
     },
     {
-      "epoch": 6.62,
+      "epoch": 6.57,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 0.3207,
+      "loss": 0.3029,
       "step": 20
     },
     {
-      "epoch": 6.92,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.25299549102783203,
-      "eval_runtime": 3.0094,
-      "eval_samples_per_second": 14.953,
-      "eval_steps_per_second": 0.665,
+      "epoch": 6.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.04919619858264923,
+      "eval_runtime": 3.3837,
+      "eval_samples_per_second": 13.89,
+      "eval_steps_per_second": 0.591,
       "step": 21
     },
     {
-      "epoch": 7.92,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.22416652739048004,
-      "eval_runtime": 3.3422,
-      "eval_samples_per_second": 13.464,
-      "eval_steps_per_second": 0.598,
+      "epoch": 7.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.05591897666454315,
+      "eval_runtime": 3.0774,
+      "eval_samples_per_second": 15.273,
+      "eval_steps_per_second": 0.65,
       "step": 24
     },
     {
-      "epoch": 8.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20596002042293549,
-      "eval_runtime": 3.2888,
-      "eval_samples_per_second": 13.683,
-      "eval_steps_per_second": 0.608,
+      "epoch": 8.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.0664142519235611,
+      "eval_runtime": 3.4113,
+      "eval_samples_per_second": 13.778,
+      "eval_steps_per_second": 0.586,
       "step": 27
     },
     {
-      "epoch": 9.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.18088287115097046,
-      "eval_runtime": 3.0438,
-      "eval_samples_per_second": 14.784,
-      "eval_steps_per_second": 0.657,
+      "epoch": 9.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.06434692442417145,
+      "eval_runtime": 3.3792,
+      "eval_samples_per_second": 13.908,
+      "eval_steps_per_second": 0.592,
       "step": 30
     },
     {
-      "epoch": 10.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20699164271354675,
-      "eval_runtime": 3.347,
-      "eval_samples_per_second": 13.445,
-      "eval_steps_per_second": 0.598,
+      "epoch": 10.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.05577840283513069,
+      "eval_runtime": 3.3341,
+      "eval_samples_per_second": 14.097,
+      "eval_steps_per_second": 0.6,
       "step": 33
     },
     {
-      "epoch": 11.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.1999157965183258,
-      "eval_runtime": 3.2411,
-      "eval_samples_per_second": 13.884,
-      "eval_steps_per_second": 0.617,
+      "epoch": 11.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.03647231683135033,
+      "eval_runtime": 3.076,
+      "eval_samples_per_second": 15.28,
+      "eval_steps_per_second": 0.65,
       "step": 36
     },
     {
-      "epoch": 12.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20134444534778595,
-      "eval_runtime": 3.001,
-      "eval_samples_per_second": 14.995,
-      "eval_steps_per_second": 0.666,
+      "epoch": 12.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.043758317828178406,
+      "eval_runtime": 3.3173,
+      "eval_samples_per_second": 14.168,
+      "eval_steps_per_second": 0.603,
       "step": 39
     },
     {
-      "epoch": 13.31,
+      "epoch": 13.29,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.2066,
+      "loss": 0.2212,
       "step": 40
     },
     {
-      "epoch": 13.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.2026776820421219,
-      "eval_runtime": 3.2971,
-      "eval_samples_per_second": 13.648,
-      "eval_steps_per_second": 0.607,
+      "epoch": 13.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.045602452009916306,
+      "eval_runtime": 3.3351,
+      "eval_samples_per_second": 14.093,
+      "eval_steps_per_second": 0.6,
       "step": 42
     },
     {
-      "epoch": 14.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.1808823049068451,
-      "eval_runtime": 3.3018,
-      "eval_samples_per_second": 13.629,
-      "eval_steps_per_second": 0.606,
+      "epoch": 14.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.040249306708574295,
+      "eval_runtime": 3.0799,
+      "eval_samples_per_second": 15.26,
+      "eval_steps_per_second": 0.649,
       "step": 45
     },
     {
-      "epoch": 15.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.16574886441230774,
-      "eval_runtime": 3.1007,
-      "eval_samples_per_second": 14.513,
-      "eval_steps_per_second": 0.645,
+      "epoch": 15.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.03505050763487816,
+      "eval_runtime": 3.2192,
+      "eval_samples_per_second": 14.6,
+      "eval_steps_per_second": 0.621,
       "step": 48
     },
     {
-      "epoch": 16.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.17283517122268677,
-      "eval_runtime": 3.305,
-      "eval_samples_per_second": 13.616,
-      "eval_steps_per_second": 0.605,
+      "epoch": 16.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.03594408929347992,
+      "eval_runtime": 3.2905,
+      "eval_samples_per_second": 14.284,
+      "eval_steps_per_second": 0.608,
       "step": 51
     },
     {
-      "epoch": 17.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20132465660572052,
-      "eval_runtime": 3.258,
-      "eval_samples_per_second": 13.812,
-      "eval_steps_per_second": 0.614,
+      "epoch": 17.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.04266700521111488,
+      "eval_runtime": 3.4643,
+      "eval_samples_per_second": 13.567,
+      "eval_steps_per_second": 0.577,
       "step": 54
     },
     {
-      "epoch": 18.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.22262603044509888,
-      "eval_runtime": 3.0138,
-      "eval_samples_per_second": 14.931,
-      "eval_steps_per_second": 0.664,
+      "epoch": 18.86,
+      "eval_accuracy": 0.9574468085106383,
+      "eval_loss": 0.04897385463118553,
+      "eval_runtime": 3.0853,
+      "eval_samples_per_second": 15.234,
+      "eval_steps_per_second": 0.648,
       "step": 57
     },
     {
-      "epoch": 19.92,
+      "epoch": 19.86,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.1894,
+      "loss": 0.186,
       "step": 60
     },
     {
-      "epoch": 19.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20911100506782532,
-      "eval_runtime": 3.2366,
-      "eval_samples_per_second": 13.903,
-      "eval_steps_per_second": 0.618,
+      "epoch": 19.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.039550021290779114,
+      "eval_runtime": 3.1131,
+      "eval_samples_per_second": 15.097,
+      "eval_steps_per_second": 0.642,
       "step": 60
     },
     {
-      "epoch": 20.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.19400011003017426,
-      "eval_runtime": 3.2491,
-      "eval_samples_per_second": 13.85,
-      "eval_steps_per_second": 0.616,
+      "epoch": 20.86,
+      "eval_accuracy": 0.9787234042553191,
+      "eval_loss": 0.029103364795446396,
+      "eval_runtime": 3.3495,
+      "eval_samples_per_second": 14.032,
+      "eval_steps_per_second": 0.597,
       "step": 63
     },
     {
-      "epoch": 21.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.19758270680904388,
-      "eval_runtime": 3.0406,
-      "eval_samples_per_second": 14.8,
-      "eval_steps_per_second": 0.658,
+      "epoch": 21.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.015161500312387943,
+      "eval_runtime": 3.3942,
+      "eval_samples_per_second": 13.847,
+      "eval_steps_per_second": 0.589,
       "step": 66
     },
     {
-      "epoch": 22.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.22320419549942017,
-      "eval_runtime": 3.2519,
-      "eval_samples_per_second": 13.838,
-      "eval_steps_per_second": 0.615,
+      "epoch": 22.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.014185278676450253,
+      "eval_runtime": 3.0833,
+      "eval_samples_per_second": 15.244,
+      "eval_steps_per_second": 0.649,
       "step": 69
     },
     {
-      "epoch": 23.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.2381269484758377,
-      "eval_runtime": 3.3206,
-      "eval_samples_per_second": 13.552,
-      "eval_steps_per_second": 0.602,
+      "epoch": 23.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.01784444786608219,
+      "eval_runtime": 3.3662,
+      "eval_samples_per_second": 13.962,
+      "eval_steps_per_second": 0.594,
       "step": 72
     },
     {
-      "epoch": 24.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.24054782092571259,
-      "eval_runtime": 2.9599,
-      "eval_samples_per_second": 15.203,
-      "eval_steps_per_second": 0.676,
+      "epoch": 24.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.017597978934645653,
+      "eval_runtime": 3.4018,
+      "eval_samples_per_second": 13.816,
+      "eval_steps_per_second": 0.588,
       "step": 75
     },
     {
-      "epoch": 25.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.224700927734375,
-      "eval_runtime": 3.266,
-      "eval_samples_per_second": 13.778,
-      "eval_steps_per_second": 0.612,
+      "epoch": 25.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.015114851295948029,
+      "eval_runtime": 3.1308,
+      "eval_samples_per_second": 15.012,
+      "eval_steps_per_second": 0.639,
       "step": 78
     },
     {
-      "epoch": 26.62,
+      "epoch": 26.57,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.1713,
+      "loss": 0.1751,
       "step": 80
     },
     {
-      "epoch": 26.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.18945971131324768,
-      "eval_runtime": 3.2466,
-      "eval_samples_per_second": 13.861,
-      "eval_steps_per_second": 0.616,
+      "epoch": 26.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.011016697622835636,
+      "eval_runtime": 3.366,
+      "eval_samples_per_second": 13.963,
+      "eval_steps_per_second": 0.594,
       "step": 81
     },
     {
-      "epoch": 27.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.18361297249794006,
-      "eval_runtime": 2.9994,
-      "eval_samples_per_second": 15.003,
-      "eval_steps_per_second": 0.667,
+      "epoch": 27.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.012051686644554138,
+      "eval_runtime": 3.3303,
+      "eval_samples_per_second": 14.113,
+      "eval_steps_per_second": 0.601,
       "step": 84
     },
     {
-      "epoch": 28.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.19853755831718445,
-      "eval_runtime": 3.2778,
-      "eval_samples_per_second": 13.729,
-      "eval_steps_per_second": 0.61,
+      "epoch": 28.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.01581379771232605,
+      "eval_runtime": 3.3324,
+      "eval_samples_per_second": 14.104,
+      "eval_steps_per_second": 0.6,
       "step": 87
     },
     {
-      "epoch": 29.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.2126913219690323,
-      "eval_runtime": 3.2397,
-      "eval_samples_per_second": 13.89,
-      "eval_steps_per_second": 0.617,
+      "epoch": 29.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.02497912012040615,
+      "eval_runtime": 3.1285,
+      "eval_samples_per_second": 15.023,
+      "eval_steps_per_second": 0.639,
       "step": 90
     },
     {
-      "epoch": 30.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20982736349105835,
-      "eval_runtime": 2.9929,
-      "eval_samples_per_second": 15.035,
-      "eval_steps_per_second": 0.668,
+      "epoch": 30.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.02920253947377205,
+      "eval_runtime": 3.3575,
+      "eval_samples_per_second": 13.999,
+      "eval_steps_per_second": 0.596,
       "step": 93
     },
     {
-      "epoch": 31.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20032857358455658,
-      "eval_runtime": 3.2633,
-      "eval_samples_per_second": 13.79,
-      "eval_steps_per_second": 0.613,
+      "epoch": 31.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.025970185175538063,
+      "eval_runtime": 3.3707,
+      "eval_samples_per_second": 13.944,
+      "eval_steps_per_second": 0.593,
       "step": 96
     },
     {
-      "epoch": 32.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.18492181599140167,
-      "eval_runtime": 3.3119,
-      "eval_samples_per_second": 13.588,
-      "eval_steps_per_second": 0.604,
+      "epoch": 32.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.020575150847434998,
+      "eval_runtime": 3.152,
+      "eval_samples_per_second": 14.911,
+      "eval_steps_per_second": 0.635,
       "step": 99
     },
     {
-      "epoch": 33.31,
+      "epoch": 33.29,
       "learning_rate": 9.259259259259259e-06,
-      "loss": 0.1428,
+      "loss": 0.1614,
       "step": 100
     },
     {
-      "epoch": 33.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.1842741221189499,
-      "eval_runtime": 2.9406,
-      "eval_samples_per_second": 15.303,
-      "eval_steps_per_second": 0.68,
+      "epoch": 33.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.018124328926205635,
+      "eval_runtime": 3.4042,
+      "eval_samples_per_second": 13.807,
+      "eval_steps_per_second": 0.588,
       "step": 102
     },
     {
-      "epoch": 34.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.1899583339691162,
-      "eval_runtime": 3.2029,
-      "eval_samples_per_second": 14.05,
-      "eval_steps_per_second": 0.624,
+      "epoch": 34.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.017009548842906952,
+      "eval_runtime": 3.4107,
+      "eval_samples_per_second": 13.78,
+      "eval_steps_per_second": 0.586,
       "step": 105
     },
     {
-      "epoch": 35.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.19720764458179474,
-      "eval_runtime": 3.2124,
-      "eval_samples_per_second": 14.008,
-      "eval_steps_per_second": 0.623,
+      "epoch": 35.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.0170438252389431,
+      "eval_runtime": 3.0972,
+      "eval_samples_per_second": 15.175,
+      "eval_steps_per_second": 0.646,
       "step": 108
     },
     {
-      "epoch": 36.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20228949189186096,
-      "eval_runtime": 2.9912,
-      "eval_samples_per_second": 15.044,
-      "eval_steps_per_second": 0.669,
+      "epoch": 36.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.017739199101924896,
+      "eval_runtime": 3.4084,
+      "eval_samples_per_second": 13.789,
+      "eval_steps_per_second": 0.587,
       "step": 111
     },
     {
-      "epoch": 37.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20600810647010803,
-      "eval_runtime": 3.269,
-      "eval_samples_per_second": 13.766,
-      "eval_steps_per_second": 0.612,
+      "epoch": 37.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.018777411431074142,
+      "eval_runtime": 3.4191,
+      "eval_samples_per_second": 13.746,
+      "eval_steps_per_second": 0.585,
       "step": 114
     },
     {
-      "epoch": 38.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.20927244424819946,
-      "eval_runtime": 3.214,
-      "eval_samples_per_second": 14.001,
-      "eval_steps_per_second": 0.622,
+      "epoch": 38.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.01889665052294731,
+      "eval_runtime": 3.0931,
+      "eval_samples_per_second": 15.195,
+      "eval_steps_per_second": 0.647,
       "step": 117
     },
     {
-      "epoch": 39.92,
+      "epoch": 39.86,
       "learning_rate": 0.0,
-      "loss": 0.1443,
+      "loss": 0.1483,
       "step": 120
     },
     {
-      "epoch": 39.92,
-      "eval_accuracy": 0.9555555555555556,
-      "eval_loss": 0.21063438057899475,
-      "eval_runtime": 3.0071,
-      "eval_samples_per_second": 14.965,
-      "eval_steps_per_second": 0.665,
+      "epoch": 39.86,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.018766822293400764,
+      "eval_runtime": 3.369,
+      "eval_samples_per_second": 13.951,
+      "eval_steps_per_second": 0.594,
       "step": 120
     },
     {
-      "epoch": 39.92,
+      "epoch": 39.86,
       "step": 120,
-      "total_flos": 3.992815176520827e+17,
-      "train_loss": 0.1958512246608734,
-      "train_runtime": 1830.7134,
-      "train_samples_per_second": 8.783,
-      "train_steps_per_second": 0.066
+      "total_flos": 4.138238964234977e+17,
+      "train_loss": 0.19912103215853375,
+      "train_runtime": 1848.5966,
+      "train_samples_per_second": 9.023,
+      "train_steps_per_second": 0.065
     }
   ],
   "max_steps": 120,
   "num_train_epochs": 40,
-  "total_flos": 3.992815176520827e+17,
+  "total_flos": 4.138238964234977e+17,
   "trial_name": null,
   "trial_params": null
 }