End of training

all_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 50.0,
-    "eval_loss": 6389.54150390625,
-    "eval_runtime": 0.6156,
-    "eval_samples_per_second": 12.995,
-    "eval_steps_per_second": 1.624,
-    "train_loss": 6297.2521484375,
-    "train_runtime": 886.6987,
-    "train_samples_per_second": 1.804,
-    "train_steps_per_second": 0.226
+    "eval_loss": 6388.880859375,
+    "eval_runtime": 0.6142,
+    "eval_samples_per_second": 13.026,
+    "eval_steps_per_second": 1.628,
+    "train_loss": 6407.14462890625,
+    "train_runtime": 869.7842,
+    "train_samples_per_second": 1.84,
+    "train_steps_per_second": 0.23
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 50.0,
-    "eval_loss": 6389.54150390625,
-    "eval_runtime": 0.6156,
-    "eval_samples_per_second": 12.995,
-    "eval_steps_per_second": 1.624
+    "eval_loss": 6388.880859375,
+    "eval_runtime": 0.6142,
+    "eval_samples_per_second": 13.026,
+    "eval_steps_per_second": 1.628
 }

runs/Jul20_11-08-08_amy-2-gpu/events.out.tfevents.1689852185.amy-2-gpu.149744.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4afae3a6e0edf71618c4506307a2497aa0aef78b9dde5fcfdaed93ac74aacf01
+size 359

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 50.0,
-    "train_loss": 6297.2521484375,
-    "train_runtime": 886.6987,
-    "train_samples_per_second": 1.804,
-    "train_steps_per_second": 0.226
+    "train_loss": 6407.14462890625,
+    "train_runtime": 869.7842,
+    "train_samples_per_second": 1.84,
+    "train_steps_per_second": 0.23
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 6389.54150390625,
-  "best_model_checkpoint": "./coco_outputs/checkpoint-196",
+  "best_metric": 6388.880859375,
+  "best_model_checkpoint": "./coco_outputs/checkpoint-200",
   "epoch": 50.0,
   "global_step": 200,
   "is_hyper_param_search": false,
@@ -9,532 +9,532 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_loss": 6398.52734375,
-      "eval_runtime": 0.5289,
-      "eval_samples_per_second": 15.126,
-      "eval_steps_per_second": 1.891,
+      "eval_loss": 6398.4208984375,
+      "eval_runtime": 0.5249,
+      "eval_samples_per_second": 15.242,
+      "eval_steps_per_second": 1.905,
       "step": 4
     },
     {
       "epoch": 2.0,
-      "eval_loss": 6397.9267578125,
-      "eval_runtime": 0.5329,
-      "eval_samples_per_second": 15.013,
-      "eval_steps_per_second": 1.877,
+      "eval_loss": 6397.8583984375,
+      "eval_runtime": 0.5335,
+      "eval_samples_per_second": 14.995,
+      "eval_steps_per_second": 1.874,
       "step": 8
     },
     {
       "epoch": 2.5,
       "learning_rate": 1.9e-05,
-      "loss": 6175.8695,
+      "loss": 6370.2805,
       "step": 10
     },
     {
       "epoch": 3.0,
-      "eval_loss": 6397.248046875,
-      "eval_runtime": 0.5325,
-      "eval_samples_per_second": 15.023,
-      "eval_steps_per_second": 1.878,
+      "eval_loss": 6397.47900390625,
+      "eval_runtime": 0.5389,
+      "eval_samples_per_second": 14.844,
+      "eval_steps_per_second": 1.855,
       "step": 12
     },
     {
       "epoch": 4.0,
-      "eval_loss": 6396.56494140625,
-      "eval_runtime": 0.5866,
-      "eval_samples_per_second": 13.638,
-      "eval_steps_per_second": 1.705,
+      "eval_loss": 6396.83203125,
+      "eval_runtime": 0.5405,
+      "eval_samples_per_second": 14.802,
+      "eval_steps_per_second": 1.85,
       "step": 16
     },
     {
       "epoch": 5.0,
       "learning_rate": 1.8e-05,
-      "loss": 6292.1676,
+      "loss": 6424.3547,
       "step": 20
     },
     {
       "epoch": 5.0,
-      "eval_loss": 6396.06689453125,
-      "eval_runtime": 0.5343,
-      "eval_samples_per_second": 14.974,
-      "eval_steps_per_second": 1.872,
+      "eval_loss": 6396.29296875,
+      "eval_runtime": 0.5443,
+      "eval_samples_per_second": 14.697,
+      "eval_steps_per_second": 1.837,
       "step": 20
     },
     {
       "epoch": 6.0,
-      "eval_loss": 6395.65966796875,
-      "eval_runtime": 0.5271,
-      "eval_samples_per_second": 15.178,
-      "eval_steps_per_second": 1.897,
+      "eval_loss": 6395.72314453125,
+      "eval_runtime": 0.5415,
+      "eval_samples_per_second": 14.775,
+      "eval_steps_per_second": 1.847,
       "step": 24
     },
     {
       "epoch": 7.0,
-      "eval_loss": 6395.27392578125,
-      "eval_runtime": 0.5241,
-      "eval_samples_per_second": 15.264,
-      "eval_steps_per_second": 1.908,
+      "eval_loss": 6395.1376953125,
+      "eval_runtime": 0.5498,
+      "eval_samples_per_second": 14.55,
+      "eval_steps_per_second": 1.819,
       "step": 28
     },
     {
       "epoch": 7.5,
       "learning_rate": 1.7e-05,
-      "loss": 6230.1191,
+      "loss": 6477.4051,
       "step": 30
     },
     {
       "epoch": 8.0,
-      "eval_loss": 6394.890625,
-      "eval_runtime": 0.5365,
-      "eval_samples_per_second": 14.911,
-      "eval_steps_per_second": 1.864,
+      "eval_loss": 6394.81884765625,
+      "eval_runtime": 0.5491,
+      "eval_samples_per_second": 14.568,
+      "eval_steps_per_second": 1.821,
       "step": 32
     },
     {
       "epoch": 9.0,
-      "eval_loss": 6394.66796875,
-      "eval_runtime": 0.5263,
-      "eval_samples_per_second": 15.201,
-      "eval_steps_per_second": 1.9,
+      "eval_loss": 6394.478515625,
+      "eval_runtime": 0.5553,
+      "eval_samples_per_second": 14.407,
+      "eval_steps_per_second": 1.801,
       "step": 36
     },
     {
       "epoch": 10.0,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 6089.0629,
+      "loss": 6381.9848,
       "step": 40
     },
     {
       "epoch": 10.0,
-      "eval_loss": 6394.51513671875,
-      "eval_runtime": 0.5595,
-      "eval_samples_per_second": 14.298,
-      "eval_steps_per_second": 1.787,
+      "eval_loss": 6394.22119140625,
+      "eval_runtime": 0.5529,
+      "eval_samples_per_second": 14.469,
+      "eval_steps_per_second": 1.809,
       "step": 40
     },
     {
       "epoch": 11.0,
-      "eval_loss": 6394.369140625,
-      "eval_runtime": 0.5347,
-      "eval_samples_per_second": 14.961,
-      "eval_steps_per_second": 1.87,
+      "eval_loss": 6394.06640625,
+      "eval_runtime": 0.5511,
+      "eval_samples_per_second": 14.516,
+      "eval_steps_per_second": 1.814,
       "step": 44
     },
     {
       "epoch": 12.0,
-      "eval_loss": 6394.39306640625,
-      "eval_runtime": 0.6066,
-      "eval_samples_per_second": 13.188,
-      "eval_steps_per_second": 1.648,
+      "eval_loss": 6393.92529296875,
+      "eval_runtime": 0.5545,
+      "eval_samples_per_second": 14.426,
+      "eval_steps_per_second": 1.803,
       "step": 48
     },
     {
       "epoch": 12.5,
       "learning_rate": 1.5000000000000002e-05,
-      "loss": 6266.6961,
+      "loss": 6343.784,
       "step": 50
     },
     {
       "epoch": 13.0,
-      "eval_loss": 6394.3232421875,
-      "eval_runtime": 0.5315,
-      "eval_samples_per_second": 15.053,
-      "eval_steps_per_second": 1.882,
+      "eval_loss": 6393.734375,
+      "eval_runtime": 0.552,
+      "eval_samples_per_second": 14.494,
+      "eval_steps_per_second": 1.812,
       "step": 52
     },
     {
       "epoch": 14.0,
-      "eval_loss": 6394.1572265625,
-      "eval_runtime": 0.5445,
-      "eval_samples_per_second": 14.694,
-      "eval_steps_per_second": 1.837,
+      "eval_loss": 6393.61279296875,
+      "eval_runtime": 0.5519,
+      "eval_samples_per_second": 14.496,
+      "eval_steps_per_second": 1.812,
       "step": 56
     },
     {
       "epoch": 15.0,
       "learning_rate": 1.4e-05,
-      "loss": 6383.0688,
+      "loss": 6458.8668,
       "step": 60
     },
     {
       "epoch": 15.0,
-      "eval_loss": 6394.15576171875,
-      "eval_runtime": 0.5377,
-      "eval_samples_per_second": 14.877,
-      "eval_steps_per_second": 1.86,
+      "eval_loss": 6393.43798828125,
+      "eval_runtime": 0.5498,
+      "eval_samples_per_second": 14.551,
+      "eval_steps_per_second": 1.819,
       "step": 60
     },
     {
       "epoch": 16.0,
-      "eval_loss": 6393.953125,
-      "eval_runtime": 0.5413,
-      "eval_samples_per_second": 14.779,
-      "eval_steps_per_second": 1.847,
+      "eval_loss": 6393.2392578125,
+      "eval_runtime": 0.5467,
+      "eval_samples_per_second": 14.633,
+      "eval_steps_per_second": 1.829,
       "step": 64
     },
     {
       "epoch": 17.0,
-      "eval_loss": 6393.77197265625,
-      "eval_runtime": 0.5527,
-      "eval_samples_per_second": 14.475,
-      "eval_steps_per_second": 1.809,
+      "eval_loss": 6393.10400390625,
+      "eval_runtime": 0.546,
+      "eval_samples_per_second": 14.651,
+      "eval_steps_per_second": 1.831,
       "step": 68
     },
     {
       "epoch": 17.5,
       "learning_rate": 1.3000000000000001e-05,
-      "loss": 6323.732,
+      "loss": 6414.077,
       "step": 70
     },
     {
       "epoch": 18.0,
-      "eval_loss": 6393.60498046875,
-      "eval_runtime": 0.5502,
-      "eval_samples_per_second": 14.54,
-      "eval_steps_per_second": 1.817,
+      "eval_loss": 6392.95751953125,
+      "eval_runtime": 0.5445,
+      "eval_samples_per_second": 14.693,
+      "eval_steps_per_second": 1.837,
       "step": 72
     },
     {
       "epoch": 19.0,
-      "eval_loss": 6393.5107421875,
-      "eval_runtime": 0.5415,
-      "eval_samples_per_second": 14.774,
-      "eval_steps_per_second": 1.847,
+      "eval_loss": 6392.830078125,
+      "eval_runtime": 0.553,
+      "eval_samples_per_second": 14.467,
+      "eval_steps_per_second": 1.808,
       "step": 76
     },
     {
       "epoch": 20.0,
       "learning_rate": 1.2e-05,
-      "loss": 6262.8367,
+      "loss": 6417.8516,
       "step": 80
     },
     {
       "epoch": 20.0,
-      "eval_loss": 6393.23046875,
-      "eval_runtime": 0.5438,
-      "eval_samples_per_second": 14.71,
-      "eval_steps_per_second": 1.839,
+      "eval_loss": 6392.3056640625,
+      "eval_runtime": 0.5433,
+      "eval_samples_per_second": 14.725,
+      "eval_steps_per_second": 1.841,
       "step": 80
     },
     {
       "epoch": 21.0,
-      "eval_loss": 6393.0869140625,
-      "eval_runtime": 0.5423,
-      "eval_samples_per_second": 14.753,
-      "eval_steps_per_second": 1.844,
+      "eval_loss": 6391.8310546875,
+      "eval_runtime": 0.5461,
+      "eval_samples_per_second": 14.65,
+      "eval_steps_per_second": 1.831,
       "step": 84
     },
     {
       "epoch": 22.0,
-      "eval_loss": 6392.791015625,
-      "eval_runtime": 0.5523,
-      "eval_samples_per_second": 14.486,
-      "eval_steps_per_second": 1.811,
+      "eval_loss": 6391.55322265625,
+      "eval_runtime": 0.5499,
+      "eval_samples_per_second": 14.547,
+      "eval_steps_per_second": 1.818,
       "step": 88
     },
     {
       "epoch": 22.5,
       "learning_rate": 1.1000000000000001e-05,
-      "loss": 6294.5953,
+      "loss": 6333.3547,
       "step": 90
     },
     {
       "epoch": 23.0,
-      "eval_loss": 6392.60791015625,
-      "eval_runtime": 0.5553,
-      "eval_samples_per_second": 14.407,
-      "eval_steps_per_second": 1.801,
+      "eval_loss": 6391.34033203125,
+      "eval_runtime": 0.5549,
+      "eval_samples_per_second": 14.418,
+      "eval_steps_per_second": 1.802,
       "step": 92
     },
     {
       "epoch": 24.0,
-      "eval_loss": 6392.46337890625,
-      "eval_runtime": 0.542,
-      "eval_samples_per_second": 14.759,
-      "eval_steps_per_second": 1.845,
+      "eval_loss": 6391.193359375,
+      "eval_runtime": 0.5535,
+      "eval_samples_per_second": 14.454,
+      "eval_steps_per_second": 1.807,
       "step": 96
     },
     {
       "epoch": 25.0,
       "learning_rate": 1e-05,
-      "loss": 6454.1902,
+      "loss": 6455.1539,
       "step": 100
     },
     {
       "epoch": 25.0,
-      "eval_loss": 6392.30859375,
-      "eval_runtime": 0.5462,
-      "eval_samples_per_second": 14.647,
-      "eval_steps_per_second": 1.831,
+      "eval_loss": 6390.97412109375,
+      "eval_runtime": 0.5532,
+      "eval_samples_per_second": 14.46,
+      "eval_steps_per_second": 1.808,
       "step": 100
     },
     {
       "epoch": 26.0,
-      "eval_loss": 6392.1982421875,
-      "eval_runtime": 0.5465,
-      "eval_samples_per_second": 14.64,
-      "eval_steps_per_second": 1.83,
+      "eval_loss": 6390.82275390625,
+      "eval_runtime": 0.562,
+      "eval_samples_per_second": 14.234,
+      "eval_steps_per_second": 1.779,
       "step": 104
     },
     {
       "epoch": 27.0,
-      "eval_loss": 6391.966796875,
-      "eval_runtime": 0.5565,
-      "eval_samples_per_second": 14.375,
-      "eval_steps_per_second": 1.797,
+      "eval_loss": 6390.7607421875,
+      "eval_runtime": 0.5645,
+      "eval_samples_per_second": 14.171,
+      "eval_steps_per_second": 1.771,
       "step": 108
     },
     {
       "epoch": 27.5,
       "learning_rate": 9e-06,
-      "loss": 6325.2328,
+      "loss": 6399.7898,
       "step": 110
     },
     {
       "epoch": 28.0,
-      "eval_loss": 6391.6533203125,
-      "eval_runtime": 0.5394,
-      "eval_samples_per_second": 14.833,
-      "eval_steps_per_second": 1.854,
+      "eval_loss": 6390.66552734375,
+      "eval_runtime": 0.558,
+      "eval_samples_per_second": 14.338,
+      "eval_steps_per_second": 1.792,
       "step": 112
     },
     {
       "epoch": 29.0,
-      "eval_loss": 6391.34814453125,
-      "eval_runtime": 0.5379,
-      "eval_samples_per_second": 14.872,
-      "eval_steps_per_second": 1.859,
+      "eval_loss": 6390.5859375,
+      "eval_runtime": 0.5544,
+      "eval_samples_per_second": 14.429,
+      "eval_steps_per_second": 1.804,
       "step": 116
     },
     {
       "epoch": 30.0,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 6471.3637,
+      "loss": 6410.9336,
       "step": 120
     },
     {
       "epoch": 30.0,
-      "eval_loss": 6391.04736328125,
-      "eval_runtime": 0.5998,
-      "eval_samples_per_second": 13.338,
-      "eval_steps_per_second": 1.667,
+      "eval_loss": 6390.49072265625,
+      "eval_runtime": 0.5598,
+      "eval_samples_per_second": 14.291,
+      "eval_steps_per_second": 1.786,
       "step": 120
     },
     {
       "epoch": 31.0,
-      "eval_loss": 6390.94091796875,
-      "eval_runtime": 0.5467,
-      "eval_samples_per_second": 14.632,
-      "eval_steps_per_second": 1.829,
+      "eval_loss": 6390.3388671875,
+      "eval_runtime": 0.562,
+      "eval_samples_per_second": 14.236,
+      "eval_steps_per_second": 1.779,
       "step": 124
     },
     {
       "epoch": 32.0,
-      "eval_loss": 6390.81982421875,
-      "eval_runtime": 0.5421,
-      "eval_samples_per_second": 14.759,
-      "eval_steps_per_second": 1.845,
+      "eval_loss": 6390.19775390625,
+      "eval_runtime": 0.546,
+      "eval_samples_per_second": 14.652,
+      "eval_steps_per_second": 1.831,
       "step": 128
     },
     {
       "epoch": 32.5,
       "learning_rate": 7e-06,
-      "loss": 6308.2656,
+      "loss": 6409.2,
       "step": 130
     },
     {
       "epoch": 33.0,
-      "eval_loss": 6390.673828125,
-      "eval_runtime": 0.5411,
-      "eval_samples_per_second": 14.785,
-      "eval_steps_per_second": 1.848,
+      "eval_loss": 6390.0341796875,
+      "eval_runtime": 0.5441,
+      "eval_samples_per_second": 14.704,
+      "eval_steps_per_second": 1.838,
       "step": 132
     },
     {
       "epoch": 34.0,
-      "eval_loss": 6390.6689453125,
-      "eval_runtime": 0.5495,
-      "eval_samples_per_second": 14.56,
-      "eval_steps_per_second": 1.82,
+      "eval_loss": 6389.96240234375,
+      "eval_runtime": 0.5472,
+      "eval_samples_per_second": 14.619,
+      "eval_steps_per_second": 1.827,
       "step": 136
     },
     {
       "epoch": 35.0,
       "learning_rate": 6e-06,
-      "loss": 6352.0371,
+      "loss": 6406.6211,
       "step": 140
     },
     {
       "epoch": 35.0,
-      "eval_loss": 6390.62451171875,
-      "eval_runtime": 0.5561,
-      "eval_samples_per_second": 14.385,
-      "eval_steps_per_second": 1.798,
+      "eval_loss": 6389.9111328125,
+      "eval_runtime": 0.5514,
+      "eval_samples_per_second": 14.509,
+      "eval_steps_per_second": 1.814,
       "step": 140
     },
     {
       "epoch": 36.0,
-      "eval_loss": 6390.5791015625,
-      "eval_runtime": 0.5601,
-      "eval_samples_per_second": 14.282,
-      "eval_steps_per_second": 1.785,
+      "eval_loss": 6389.6875,
+      "eval_runtime": 0.5453,
+      "eval_samples_per_second": 14.672,
+      "eval_steps_per_second": 1.834,
       "step": 144
     },
     {
       "epoch": 37.0,
-      "eval_loss": 6390.419921875,
-      "eval_runtime": 0.5447,
-      "eval_samples_per_second": 14.686,
-      "eval_steps_per_second": 1.836,
+      "eval_loss": 6389.4755859375,
+      "eval_runtime": 0.5552,
+      "eval_samples_per_second": 14.409,
+      "eval_steps_per_second": 1.801,
       "step": 148
     },
     {
       "epoch": 37.5,
       "learning_rate": 5e-06,
-      "loss": 6247.1492,
+      "loss": 6371.1539,
       "step": 150
     },
     {
       "epoch": 38.0,
-      "eval_loss": 6390.23046875,
-      "eval_runtime": 0.5405,
-      "eval_samples_per_second": 14.8,
-      "eval_steps_per_second": 1.85,
+      "eval_loss": 6389.3515625,
+      "eval_runtime": 0.5423,
+      "eval_samples_per_second": 14.751,
+      "eval_steps_per_second": 1.844,
       "step": 152
     },
     {
       "epoch": 39.0,
-      "eval_loss": 6390.07763671875,
-      "eval_runtime": 0.5527,
-      "eval_samples_per_second": 14.473,
-      "eval_steps_per_second": 1.809,
+      "eval_loss": 6389.26953125,
+      "eval_runtime": 0.5522,
+      "eval_samples_per_second": 14.487,
+      "eval_steps_per_second": 1.811,
       "step": 156
     },
     {
       "epoch": 40.0,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 6385.748,
+      "loss": 6409.1055,
       "step": 160
     },
     {
       "epoch": 40.0,
-      "eval_loss": 6389.9873046875,
-      "eval_runtime": 0.5425,
-      "eval_samples_per_second": 14.746,
-      "eval_steps_per_second": 1.843,
+      "eval_loss": 6389.24951171875,
+      "eval_runtime": 0.5572,
+      "eval_samples_per_second": 14.356,
+      "eval_steps_per_second": 1.795,
       "step": 160
     },
     {
       "epoch": 41.0,
-      "eval_loss": 6389.91845703125,
-      "eval_runtime": 0.5483,
-      "eval_samples_per_second": 14.591,
-      "eval_steps_per_second": 1.824,
+      "eval_loss": 6389.208984375,
+      "eval_runtime": 0.5378,
+      "eval_samples_per_second": 14.876,
+      "eval_steps_per_second": 1.86,
       "step": 164
     },
     {
       "epoch": 42.0,
-      "eval_loss": 6389.828125,
-      "eval_runtime": 0.5625,
-      "eval_samples_per_second": 14.221,
-      "eval_steps_per_second": 1.778,
+      "eval_loss": 6389.10986328125,
+      "eval_runtime": 0.539,
+      "eval_samples_per_second": 14.841,
+      "eval_steps_per_second": 1.855,
       "step": 168
     },
     {
       "epoch": 42.5,
       "learning_rate": 3e-06,
-      "loss": 6262.0703,
+      "loss": 6453.5285,
       "step": 170
     },
     {
       "epoch": 43.0,
-      "eval_loss": 6389.75,
-      "eval_runtime": 0.5518,
-      "eval_samples_per_second": 14.498,
-      "eval_steps_per_second": 1.812,
+      "eval_loss": 6389.04052734375,
+      "eval_runtime": 0.54,
+      "eval_samples_per_second": 14.815,
+      "eval_steps_per_second": 1.852,
       "step": 172
     },
     {
       "epoch": 44.0,
-      "eval_loss": 6389.67919921875,
-      "eval_runtime": 0.5455,
-      "eval_samples_per_second": 14.666,
-      "eval_steps_per_second": 1.833,
+      "eval_loss": 6388.99365234375,
+      "eval_runtime": 0.5364,
+      "eval_samples_per_second": 14.915,
+      "eval_steps_per_second": 1.864,
       "step": 176
     },
     {
       "epoch": 45.0,
       "learning_rate": 2.0000000000000003e-06,
-      "loss": 6256.2105,
+      "loss": 6391.1004,
       "step": 180
     },
     {
       "epoch": 45.0,
-      "eval_loss": 6389.6376953125,
-      "eval_runtime": 0.5593,
-      "eval_samples_per_second": 14.304,
-      "eval_steps_per_second": 1.788,
+      "eval_loss": 6388.9541015625,
+      "eval_runtime": 0.5495,
+      "eval_samples_per_second": 14.558,
+      "eval_steps_per_second": 1.82,
       "step": 180
     },
     {
       "epoch": 46.0,
-      "eval_loss": 6389.6201171875,
-      "eval_runtime": 0.555,
-      "eval_samples_per_second": 14.413,
-      "eval_steps_per_second": 1.802,
+      "eval_loss": 6388.923828125,
+      "eval_runtime": 0.5496,
+      "eval_samples_per_second": 14.557,
+      "eval_steps_per_second": 1.82,
       "step": 184
     },
     {
       "epoch": 47.0,
-      "eval_loss": 6389.57275390625,
-      "eval_runtime": 0.5436,
-      "eval_samples_per_second": 14.718,
-      "eval_steps_per_second": 1.84,
+      "eval_loss": 6388.90771484375,
+      "eval_runtime": 0.5379,
+      "eval_samples_per_second": 14.871,
+      "eval_steps_per_second": 1.859,
       "step": 188
     },
     {
       "epoch": 47.5,
       "learning_rate": 1.0000000000000002e-06,
-      "loss": 6353.4496,
+      "loss": 6416.6641,
       "step": 190
     },
     {
       "epoch": 48.0,
-      "eval_loss": 6389.5517578125,
-      "eval_runtime": 0.55,
-      "eval_samples_per_second": 14.545,
-      "eval_steps_per_second": 1.818,
+      "eval_loss": 6388.89111328125,
+      "eval_runtime": 0.5385,
+      "eval_samples_per_second": 14.857,
+      "eval_steps_per_second": 1.857,
       "step": 192
     },
     {
       "epoch": 49.0,
-      "eval_loss": 6389.54150390625,
-      "eval_runtime": 0.5412,
-      "eval_samples_per_second": 14.782,
-      "eval_steps_per_second": 1.848,
+      "eval_loss": 6388.8828125,
+      "eval_runtime": 0.5466,
+      "eval_samples_per_second": 14.635,
+      "eval_steps_per_second": 1.829,
       "step": 196
     },
     {
       "epoch": 50.0,
       "learning_rate": 0.0,
-      "loss": 6211.1777,
+      "loss": 6397.6828,
       "step": 200
     },
     {
       "epoch": 50.0,
-      "eval_loss": 6389.552734375,
-      "eval_runtime": 0.5488,
-      "eval_samples_per_second": 14.577,
-      "eval_steps_per_second": 1.822,
+      "eval_loss": 6388.880859375,
+      "eval_runtime": 0.5601,
+      "eval_samples_per_second": 14.283,
+      "eval_steps_per_second": 1.785,
       "step": 200
     },
     {
       "epoch": 50.0,
       "step": 200,
       "total_flos": 7.65002115072e+17,
-      "train_loss": 6297.2521484375,
-      "train_runtime": 886.6987,
-      "train_samples_per_second": 1.804,
-      "train_steps_per_second": 0.226
+      "train_loss": 6407.14462890625,
+      "train_runtime": 869.7842,
+      "train_samples_per_second": 1.84,
+      "train_steps_per_second": 0.23
     }
   ],
   "max_steps": 200,