Upload 13 files

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a9f01669d8289a4d57c8fd724bd8d9f484632beb15e26a77784f35bc45e9a1d
+oid sha256:6a262b856f4c1a0cfbfef00ac9958b691d0e686ee2df247e478fa6751938b49f
 size 1294929093

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9c96ed2870ed3d3db5083ac45980f620d8f029e2b2e652cb90be6bb9b74dd4f
+oid sha256:d0a12747c5efd2c88b7589f2241290038902ee34fb2c4256889a3fcf38815d3f
 size 647680813

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71fe3c7dba1972246838a50915d7132ef63e8e062fe45107ab37a8d1d76c54e5
+oid sha256:5a5b85b73858071bd61066c1dc15a99fbacdd3a556c9b57080cd2fcac34e566f
 size 14575

trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 9.478559583037194,
+  "epoch": 10.988739357319417,
   "global_step": 10000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
@@ -10,1350 +10,1350 @@
     {
       "epoch": 0.05,
       "learning_rate": 1.6666666666666667e-06,
-      "loss": 4.2334,
+      "loss": 4.211,
       "step": 50
     },
     {
-      "epoch": 0.09,
+      "epoch": 0.11,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 3.8483,
+      "loss": 3.7956,
       "step": 100
     },
     {
-      "epoch": 0.14,
+      "epoch": 0.16,
       "learning_rate": 5e-06,
-      "loss": 3.7169,
+      "loss": 3.6773,
       "step": 150
     },
     {
-      "epoch": 0.19,
+      "epoch": 0.22,
       "learning_rate": 6.666666666666667e-06,
-      "loss": 3.6375,
+      "loss": 3.6206,
       "step": 200
     },
     {
-      "epoch": 0.24,
+      "epoch": 0.27,
       "learning_rate": 8.333333333333334e-06,
-      "loss": 3.6413,
+      "loss": 3.6346,
       "step": 250
     },
     {
-      "epoch": 0.28,
+      "epoch": 0.33,
       "learning_rate": 1e-05,
-      "loss": 3.5774,
+      "loss": 3.5553,
       "step": 300
     },
     {
-      "epoch": 0.33,
+      "epoch": 0.38,
       "learning_rate": 1.1666666666666668e-05,
-      "loss": 3.5074,
+      "loss": 3.5336,
       "step": 350
     },
     {
-      "epoch": 0.38,
+      "epoch": 0.44,
       "learning_rate": 1.3333333333333333e-05,
-      "loss": 3.5283,
+      "loss": 3.4773,
       "step": 400
     },
     {
-      "epoch": 0.43,
+      "epoch": 0.49,
       "learning_rate": 1.5e-05,
-      "loss": 3.5295,
+      "loss": 3.5169,
       "step": 450
     },
     {
-      "epoch": 0.47,
+      "epoch": 0.55,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 3.5113,
+      "loss": 3.4958,
       "step": 500
     },
     {
-      "epoch": 0.52,
+      "epoch": 0.6,
       "learning_rate": 1.8333333333333333e-05,
-      "loss": 3.518,
+      "loss": 3.481,
       "step": 550
     },
     {
-      "epoch": 0.57,
+      "epoch": 0.66,
       "learning_rate": 2e-05,
-      "loss": 3.505,
+      "loss": 3.5022,
       "step": 600
     },
     {
-      "epoch": 0.62,
+      "epoch": 0.71,
       "learning_rate": 2.1666666666666667e-05,
-      "loss": 3.4983,
+      "loss": 3.4659,
       "step": 650
     },
     {
-      "epoch": 0.66,
+      "epoch": 0.77,
       "learning_rate": 2.3333333333333336e-05,
-      "loss": 3.5032,
+      "loss": 3.4671,
       "step": 700
     },
     {
-      "epoch": 0.71,
+      "epoch": 0.82,
       "learning_rate": 2.5e-05,
-      "loss": 3.5299,
+      "loss": 3.4435,
       "step": 750
     },
     {
-      "epoch": 0.71,
-      "eval_loss": 3.366128444671631,
-      "eval_rouge2_fmeasure": 0.0621,
-      "eval_rouge2_precision": 0.0787,
-      "eval_rouge2_recall": 0.0679,
-      "eval_runtime": 2930.3941,
-      "eval_samples_per_second": 0.32,
-      "eval_steps_per_second": 0.32,
+      "epoch": 0.82,
+      "eval_loss": 3.368271827697754,
+      "eval_rouge2_fmeasure": 0.063,
+      "eval_rouge2_precision": 0.0759,
+      "eval_rouge2_recall": 0.0713,
+      "eval_runtime": 3297.6177,
+      "eval_samples_per_second": 0.276,
+      "eval_steps_per_second": 0.276,
       "step": 750
     },
     {
-      "epoch": 0.76,
+      "epoch": 0.88,
       "learning_rate": 2.6666666666666667e-05,
-      "loss": 3.4298,
+      "loss": 3.4261,
       "step": 800
     },
     {
-      "epoch": 0.81,
+      "epoch": 0.93,
       "learning_rate": 2.8333333333333335e-05,
-      "loss": 3.4886,
+      "loss": 3.4547,
       "step": 850
     },
     {
-      "epoch": 0.85,
+      "epoch": 0.99,
       "learning_rate": 3e-05,
-      "loss": 3.4473,
+      "loss": 3.4793,
       "step": 900
     },
     {
-      "epoch": 0.9,
+      "epoch": 1.04,
       "learning_rate": 3.1666666666666666e-05,
-      "loss": 3.4084,
+      "loss": 3.357,
       "step": 950
     },
     {
-      "epoch": 0.95,
+      "epoch": 1.1,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 3.4568,
+      "loss": 3.2581,
       "step": 1000
     },
     {
-      "epoch": 1.0,
+      "epoch": 1.15,
       "learning_rate": 3.5e-05,
-      "loss": 3.4452,
+      "loss": 3.2646,
       "step": 1050
     },
     {
-      "epoch": 1.04,
+      "epoch": 1.21,
       "learning_rate": 3.6666666666666666e-05,
-      "loss": 3.2616,
+      "loss": 3.2718,
       "step": 1100
     },
     {
-      "epoch": 1.09,
+      "epoch": 1.26,
       "learning_rate": 3.8333333333333334e-05,
-      "loss": 3.2386,
+      "loss": 3.2793,
       "step": 1150
     },
     {
-      "epoch": 1.14,
+      "epoch": 1.32,
       "learning_rate": 4e-05,
-      "loss": 3.3304,
+      "loss": 3.3191,
       "step": 1200
     },
     {
-      "epoch": 1.18,
+      "epoch": 1.37,
       "learning_rate": 4.166666666666667e-05,
-      "loss": 3.2993,
+      "loss": 3.1993,
       "step": 1250
     },
     {
-      "epoch": 1.23,
+      "epoch": 1.43,
       "learning_rate": 4.3333333333333334e-05,
-      "loss": 3.2656,
+      "loss": 3.3049,
       "step": 1300
     },
     {
-      "epoch": 1.28,
+      "epoch": 1.48,
       "learning_rate": 4.5e-05,
-      "loss": 3.2852,
+      "loss": 3.3176,
       "step": 1350
     },
     {
-      "epoch": 1.33,
+      "epoch": 1.54,
       "learning_rate": 4.666666666666667e-05,
-      "loss": 3.2627,
+      "loss": 3.2733,
       "step": 1400
     },
     {
-      "epoch": 1.37,
+      "epoch": 1.59,
       "learning_rate": 4.8333333333333334e-05,
-      "loss": 3.2663,
+      "loss": 3.2297,
       "step": 1450
     },
     {
-      "epoch": 1.42,
+      "epoch": 1.65,
       "learning_rate": 5e-05,
-      "loss": 3.2872,
+      "loss": 3.2481,
       "step": 1500
     },
     {
-      "epoch": 1.42,
-      "eval_loss": 3.289588689804077,
-      "eval_rouge2_fmeasure": 0.0624,
-      "eval_rouge2_precision": 0.0779,
-      "eval_rouge2_recall": 0.0697,
-      "eval_runtime": 2994.1603,
-      "eval_samples_per_second": 0.314,
-      "eval_steps_per_second": 0.314,
+      "epoch": 1.65,
+      "eval_loss": 3.3231685161590576,
+      "eval_rouge2_fmeasure": 0.0619,
+      "eval_rouge2_precision": 0.0734,
+      "eval_rouge2_recall": 0.0716,
+      "eval_runtime": 3492.6262,
+      "eval_samples_per_second": 0.261,
+      "eval_steps_per_second": 0.261,
       "step": 1500
     },
     {
-      "epoch": 1.47,
+      "epoch": 1.7,
       "learning_rate": 4.970588235294118e-05,
-      "loss": 3.3058,
+      "loss": 3.3217,
       "step": 1550
     },
     {
-      "epoch": 1.52,
+      "epoch": 1.76,
       "learning_rate": 4.9411764705882355e-05,
-      "loss": 3.2395,
+      "loss": 3.2683,
       "step": 1600
     },
     {
-      "epoch": 1.56,
+      "epoch": 1.81,
       "learning_rate": 4.911764705882353e-05,
-      "loss": 3.2136,
+      "loss": 3.2523,
       "step": 1650
     },
     {
-      "epoch": 1.61,
+      "epoch": 1.87,
       "learning_rate": 4.882352941176471e-05,
-      "loss": 3.2615,
+      "loss": 3.1994,
       "step": 1700
     },
     {
-      "epoch": 1.66,
+      "epoch": 1.92,
       "learning_rate": 4.8529411764705885e-05,
-      "loss": 3.2615,
+      "loss": 3.2815,
       "step": 1750
     },
     {
-      "epoch": 1.71,
+      "epoch": 1.98,
       "learning_rate": 4.823529411764706e-05,
-      "loss": 3.2535,
+      "loss": 3.2126,
       "step": 1800
     },
     {
-      "epoch": 1.75,
+      "epoch": 2.03,
       "learning_rate": 4.794117647058824e-05,
-      "loss": 3.2741,
+      "loss": 3.0612,
       "step": 1850
     },
     {
-      "epoch": 1.8,
+      "epoch": 2.09,
       "learning_rate": 4.7647058823529414e-05,
-      "loss": 3.2633,
+      "loss": 2.9348,
       "step": 1900
     },
     {
-      "epoch": 1.85,
+      "epoch": 2.14,
       "learning_rate": 4.735294117647059e-05,
-      "loss": 3.2387,
+      "loss": 2.9331,
       "step": 1950
     },
     {
-      "epoch": 1.9,
+      "epoch": 2.2,
       "learning_rate": 4.705882352941177e-05,
-      "loss": 3.2711,
+      "loss": 2.9507,
       "step": 2000
     },
     {
-      "epoch": 1.94,
+      "epoch": 2.25,
       "learning_rate": 4.6764705882352944e-05,
-      "loss": 3.2466,
+      "loss": 2.9401,
       "step": 2050
     },
     {
-      "epoch": 1.99,
+      "epoch": 2.31,
       "learning_rate": 4.647058823529412e-05,
-      "loss": 3.2105,
+      "loss": 2.9139,
       "step": 2100
     },
     {
-      "epoch": 2.04,
+      "epoch": 2.36,
       "learning_rate": 4.61764705882353e-05,
-      "loss": 2.9351,
+      "loss": 2.9651,
       "step": 2150
     },
     {
-      "epoch": 2.09,
+      "epoch": 2.42,
       "learning_rate": 4.588235294117647e-05,
-      "loss": 2.9632,
+      "loss": 2.9302,
       "step": 2200
     },
     {
-      "epoch": 2.13,
+      "epoch": 2.47,
       "learning_rate": 4.558823529411765e-05,
-      "loss": 2.9587,
+      "loss": 2.9575,
       "step": 2250
     },
     {
-      "epoch": 2.13,
-      "eval_loss": 3.2455530166625977,
-      "eval_rouge2_fmeasure": 0.0679,
-      "eval_rouge2_precision": 0.0841,
-      "eval_rouge2_recall": 0.0775,
-      "eval_runtime": 3185.1079,
-      "eval_samples_per_second": 0.295,
-      "eval_steps_per_second": 0.295,
+      "epoch": 2.47,
+      "eval_loss": 3.3035736083984375,
+      "eval_rouge2_fmeasure": 0.0694,
+      "eval_rouge2_precision": 0.0806,
+      "eval_rouge2_recall": 0.0809,
+      "eval_runtime": 3378.6348,
+      "eval_samples_per_second": 0.269,
+      "eval_steps_per_second": 0.269,
       "step": 2250
     },
     {
-      "epoch": 2.18,
+      "epoch": 2.53,
       "learning_rate": 4.5294117647058826e-05,
-      "loss": 2.9037,
+      "loss": 2.9562,
       "step": 2300
     },
     {
-      "epoch": 2.23,
+      "epoch": 2.58,
       "learning_rate": 4.5e-05,
-      "loss": 2.9636,
+      "loss": 2.9469,
       "step": 2350
     },
     {
-      "epoch": 2.27,
+      "epoch": 2.64,
       "learning_rate": 4.470588235294118e-05,
-      "loss": 2.915,
+      "loss": 2.9122,
       "step": 2400
     },
     {
-      "epoch": 2.32,
+      "epoch": 2.69,
       "learning_rate": 4.4411764705882356e-05,
-      "loss": 2.8962,
+      "loss": 2.9922,
       "step": 2450
     },
     {
-      "epoch": 2.37,
+      "epoch": 2.75,
       "learning_rate": 4.411764705882353e-05,
-      "loss": 2.8742,
+      "loss": 3.0068,
       "step": 2500
     },
     {
-      "epoch": 2.42,
+      "epoch": 2.8,
       "learning_rate": 4.382352941176471e-05,
-      "loss": 2.9218,
+      "loss": 2.9438,
       "step": 2550
     },
     {
-      "epoch": 2.46,
+      "epoch": 2.86,
       "learning_rate": 4.3529411764705885e-05,
-      "loss": 2.9232,
+      "loss": 2.9429,
       "step": 2600
     },
     {
-      "epoch": 2.51,
+      "epoch": 2.91,
       "learning_rate": 4.323529411764706e-05,
-      "loss": 2.9556,
+      "loss": 2.9085,
       "step": 2650
     },
     {
-      "epoch": 2.56,
+      "epoch": 2.97,
       "learning_rate": 4.294117647058823e-05,
-      "loss": 2.8983,
+      "loss": 2.9765,
       "step": 2700
     },
     {
-      "epoch": 2.61,
+      "epoch": 3.02,
       "learning_rate": 4.2647058823529415e-05,
-      "loss": 3.0018,
+      "loss": 2.7701,
       "step": 2750
     },
     {
-      "epoch": 2.65,
+      "epoch": 3.08,
       "learning_rate": 4.235294117647059e-05,
-      "loss": 2.9017,
+      "loss": 2.5671,
       "step": 2800
     },
     {
-      "epoch": 2.7,
+      "epoch": 3.13,
       "learning_rate": 4.205882352941177e-05,
-      "loss": 2.9353,
+      "loss": 2.6589,
       "step": 2850
     },
     {
-      "epoch": 2.75,
+      "epoch": 3.19,
       "learning_rate": 4.1764705882352944e-05,
-      "loss": 2.991,
+      "loss": 2.644,
       "step": 2900
     },
     {
-      "epoch": 2.8,
+      "epoch": 3.24,
       "learning_rate": 4.147058823529412e-05,
-      "loss": 2.9332,
+      "loss": 2.6558,
       "step": 2950
     },
     {
-      "epoch": 2.84,
+      "epoch": 3.3,
       "learning_rate": 4.11764705882353e-05,
-      "loss": 2.9439,
+      "loss": 2.7165,
       "step": 3000
     },
     {
-      "epoch": 2.84,
-      "eval_loss": 3.1968302726745605,
-      "eval_rouge2_fmeasure": 0.0742,
-      "eval_rouge2_precision": 0.0906,
-      "eval_rouge2_recall": 0.0824,
-      "eval_runtime": 2977.5543,
-      "eval_samples_per_second": 0.315,
-      "eval_steps_per_second": 0.315,
+      "epoch": 3.3,
+      "eval_loss": 3.2863686084747314,
+      "eval_rouge2_fmeasure": 0.0697,
+      "eval_rouge2_precision": 0.0803,
+      "eval_rouge2_recall": 0.0828,
+      "eval_runtime": 3409.6518,
+      "eval_samples_per_second": 0.267,
+      "eval_steps_per_second": 0.267,
       "step": 3000
     },
     {
-      "epoch": 2.89,
+      "epoch": 3.35,
       "learning_rate": 4.0882352941176474e-05,
-      "loss": 2.9403,
+      "loss": 2.6817,
       "step": 3050
     },
     {
-      "epoch": 2.94,
+      "epoch": 3.41,
       "learning_rate": 4.058823529411765e-05,
-      "loss": 2.9788,
+      "loss": 2.5758,
       "step": 3100
     },
     {
-      "epoch": 2.99,
+      "epoch": 3.46,
       "learning_rate": 4.029411764705883e-05,
-      "loss": 2.8354,
+      "loss": 2.6271,
       "step": 3150
     },
     {
-      "epoch": 3.03,
+      "epoch": 3.52,
       "learning_rate": 4e-05,
-      "loss": 2.7561,
+      "loss": 2.6614,
       "step": 3200
     },
     {
-      "epoch": 3.08,
+      "epoch": 3.57,
       "learning_rate": 3.970588235294117e-05,
-      "loss": 2.6053,
+      "loss": 2.5872,
       "step": 3250
     },
     {
-      "epoch": 3.13,
+      "epoch": 3.63,
       "learning_rate": 3.9411764705882356e-05,
-      "loss": 2.6016,
+      "loss": 2.6659,
       "step": 3300
     },
     {
-      "epoch": 3.18,
+      "epoch": 3.68,
       "learning_rate": 3.911764705882353e-05,
-      "loss": 2.6452,
+      "loss": 2.7674,
       "step": 3350
     },
     {
-      "epoch": 3.22,
+      "epoch": 3.74,
       "learning_rate": 3.882352941176471e-05,
-      "loss": 2.6891,
+      "loss": 2.6768,
       "step": 3400
     },
     {
-      "epoch": 3.27,
+      "epoch": 3.79,
       "learning_rate": 3.8529411764705886e-05,
-      "loss": 2.6705,
+      "loss": 2.6378,
       "step": 3450
     },
     {
-      "epoch": 3.32,
+      "epoch": 3.85,
       "learning_rate": 3.8235294117647055e-05,
-      "loss": 2.6486,
+      "loss": 2.6802,
       "step": 3500
     },
     {
-      "epoch": 3.36,
+      "epoch": 3.9,
       "learning_rate": 3.794117647058824e-05,
-      "loss": 2.6336,
+      "loss": 2.7035,
       "step": 3550
     },
     {
-      "epoch": 3.41,
+      "epoch": 3.96,
       "learning_rate": 3.7647058823529415e-05,
-      "loss": 2.6828,
+      "loss": 2.6615,
       "step": 3600
     },
     {
-      "epoch": 3.46,
+      "epoch": 4.01,
       "learning_rate": 3.735294117647059e-05,
-      "loss": 2.6872,
+      "loss": 2.5905,
       "step": 3650
     },
     {
-      "epoch": 3.51,
+      "epoch": 4.07,
       "learning_rate": 3.705882352941177e-05,
-      "loss": 2.6874,
+      "loss": 2.4881,
       "step": 3700
     },
     {
-      "epoch": 3.55,
+      "epoch": 4.12,
       "learning_rate": 3.6764705882352945e-05,
-      "loss": 2.6839,
+      "loss": 2.3811,
       "step": 3750
     },
     {
-      "epoch": 3.55,
-      "eval_loss": 3.22872257232666,
-      "eval_rouge2_fmeasure": 0.0729,
-      "eval_rouge2_precision": 0.0878,
-      "eval_rouge2_recall": 0.0833,
-      "eval_runtime": 3260.1882,
-      "eval_samples_per_second": 0.288,
-      "eval_steps_per_second": 0.288,
+      "epoch": 4.12,
+      "eval_loss": 3.319762706756592,
+      "eval_rouge2_fmeasure": 0.0696,
+      "eval_rouge2_precision": 0.0818,
+      "eval_rouge2_recall": 0.0816,
+      "eval_runtime": 3078.5804,
+      "eval_samples_per_second": 0.296,
+      "eval_steps_per_second": 0.296,
       "step": 3750
     },
     {
-      "epoch": 3.6,
+      "epoch": 4.18,
       "learning_rate": 3.6470588235294114e-05,
-      "loss": 2.6313,
+      "loss": 2.4033,
       "step": 3800
     },
     {
-      "epoch": 3.65,
+      "epoch": 4.23,
       "learning_rate": 3.61764705882353e-05,
-      "loss": 2.6294,
+      "loss": 2.395,
       "step": 3850
     },
     {
-      "epoch": 3.7,
+      "epoch": 4.29,
       "learning_rate": 3.5882352941176474e-05,
-      "loss": 2.6544,
+      "loss": 2.3486,
       "step": 3900
     },
     {
-      "epoch": 3.74,
+      "epoch": 4.34,
       "learning_rate": 3.558823529411765e-05,
-      "loss": 2.6728,
+      "loss": 2.4231,
       "step": 3950
     },
     {
-      "epoch": 3.79,
+      "epoch": 4.4,
       "learning_rate": 3.529411764705883e-05,
-      "loss": 2.6223,
+      "loss": 2.4483,
       "step": 4000
     },
     {
-      "epoch": 3.84,
+      "epoch": 4.45,
       "learning_rate": 3.5e-05,
-      "loss": 2.6436,
+      "loss": 2.3744,
       "step": 4050
     },
     {
-      "epoch": 3.89,
+      "epoch": 4.51,
       "learning_rate": 3.470588235294118e-05,
-      "loss": 2.6224,
+      "loss": 2.4387,
       "step": 4100
     },
     {
-      "epoch": 3.93,
+      "epoch": 4.56,
       "learning_rate": 3.441176470588236e-05,
-      "loss": 2.6629,
+      "loss": 2.4342,
       "step": 4150
     },
     {
-      "epoch": 3.98,
+      "epoch": 4.62,
       "learning_rate": 3.411764705882353e-05,
-      "loss": 2.65,
+      "loss": 2.3778,
       "step": 4200
     },
     {
-      "epoch": 4.03,
+      "epoch": 4.67,
       "learning_rate": 3.382352941176471e-05,
-      "loss": 2.5638,
+      "loss": 2.3947,
       "step": 4250
     },
     {
-      "epoch": 4.08,
+      "epoch": 4.73,
       "learning_rate": 3.352941176470588e-05,
-      "loss": 2.3744,
+      "loss": 2.4558,
       "step": 4300
     },
     {
-      "epoch": 4.12,
+      "epoch": 4.78,
       "learning_rate": 3.3235294117647056e-05,
-      "loss": 2.3195,
+      "loss": 2.455,
       "step": 4350
     },
     {
-      "epoch": 4.17,
+      "epoch": 4.83,
       "learning_rate": 3.294117647058824e-05,
-      "loss": 2.465,
+      "loss": 2.4651,
       "step": 4400
     },
     {
-      "epoch": 4.22,
+      "epoch": 4.89,
       "learning_rate": 3.2647058823529416e-05,
-      "loss": 2.4692,
+      "loss": 2.4458,
       "step": 4450
     },
     {
-      "epoch": 4.27,
+      "epoch": 4.94,
       "learning_rate": 3.235294117647059e-05,
-      "loss": 2.3452,
+      "loss": 2.4682,
       "step": 4500
     },
     {
-      "epoch": 4.27,
-      "eval_loss": 3.2970633506774902,
-      "eval_rouge2_fmeasure": 0.0732,
-      "eval_rouge2_precision": 0.0889,
-      "eval_rouge2_recall": 0.0826,
-      "eval_runtime": 3047.2227,
-      "eval_samples_per_second": 0.308,
-      "eval_steps_per_second": 0.308,
+      "epoch": 4.94,
+      "eval_loss": 3.3358490467071533,
+      "eval_rouge2_fmeasure": 0.0677,
+      "eval_rouge2_precision": 0.0789,
+      "eval_rouge2_recall": 0.0802,
+      "eval_runtime": 3340.0486,
+      "eval_samples_per_second": 0.272,
+      "eval_steps_per_second": 0.272,
       "step": 4500
     },
     {
-      "epoch": 4.31,
+      "epoch": 5.0,
       "learning_rate": 3.205882352941177e-05,
-      "loss": 2.4194,
+      "loss": 2.3945,
       "step": 4550
     },
     {
-      "epoch": 4.36,
+      "epoch": 5.05,
       "learning_rate": 3.176470588235294e-05,
-      "loss": 2.427,
+      "loss": 2.223,
       "step": 4600
     },
     {
-      "epoch": 4.41,
+      "epoch": 5.11,
       "learning_rate": 3.147058823529412e-05,
-      "loss": 2.4283,
+      "loss": 2.2142,
       "step": 4650
     },
     {
-      "epoch": 4.45,
+      "epoch": 5.16,
       "learning_rate": 3.11764705882353e-05,
-      "loss": 2.4371,
+      "loss": 2.1276,
       "step": 4700
     },
     {
-      "epoch": 4.5,
+      "epoch": 5.22,
       "learning_rate": 3.0882352941176475e-05,
-      "loss": 2.4109,
+      "loss": 2.2078,
       "step": 4750
     },
     {
-      "epoch": 4.55,
+      "epoch": 5.27,
       "learning_rate": 3.058823529411765e-05,
-      "loss": 2.4679,
+      "loss": 2.2406,
       "step": 4800
     },
     {
-      "epoch": 4.6,
+      "epoch": 5.33,
       "learning_rate": 3.0294117647058824e-05,
-      "loss": 2.3902,
+      "loss": 2.1769,
       "step": 4850
     },
     {
-      "epoch": 4.64,
+      "epoch": 5.38,
       "learning_rate": 3e-05,
-      "loss": 2.4985,
+      "loss": 2.2306,
       "step": 4900
     },
     {
-      "epoch": 4.69,
+      "epoch": 5.44,
       "learning_rate": 2.9705882352941177e-05,
-      "loss": 2.3842,
+      "loss": 2.1929,
       "step": 4950
     },
     {
-      "epoch": 4.74,
+      "epoch": 5.49,
       "learning_rate": 2.9411764705882354e-05,
-      "loss": 2.3876,
+      "loss": 2.2385,
       "step": 5000
     },
     {
-      "epoch": 4.79,
+      "epoch": 5.55,
       "learning_rate": 2.9117647058823534e-05,
-      "loss": 2.4887,
+      "loss": 2.2636,
       "step": 5050
     },
     {
-      "epoch": 4.83,
+      "epoch": 5.6,
       "learning_rate": 2.8823529411764703e-05,
-      "loss": 2.4936,
+      "loss": 2.2987,
       "step": 5100
     },
     {
-      "epoch": 4.88,
+      "epoch": 5.66,
       "learning_rate": 2.8529411764705883e-05,
-      "loss": 2.4209,
+      "loss": 2.2562,
       "step": 5150
     },
     {
-      "epoch": 4.93,
+      "epoch": 5.71,
       "learning_rate": 2.823529411764706e-05,
-      "loss": 2.4856,
+      "loss": 2.1916,
       "step": 5200
     },
     {
-      "epoch": 4.98,
+      "epoch": 5.77,
       "learning_rate": 2.7941176470588236e-05,
-      "loss": 2.465,
+      "loss": 2.1475,
       "step": 5250
     },
     {
-      "epoch": 4.98,
-      "eval_loss": 3.2449326515197754,
-      "eval_rouge2_fmeasure": 0.0773,
-      "eval_rouge2_precision": 0.0937,
-      "eval_rouge2_recall": 0.0868,
-      "eval_runtime": 3043.7989,
-      "eval_samples_per_second": 0.308,
-      "eval_steps_per_second": 0.308,
+      "epoch": 5.77,
+      "eval_loss": 3.3984901905059814,
+      "eval_rouge2_fmeasure": 0.0722,
+      "eval_rouge2_precision": 0.0839,
+      "eval_rouge2_recall": 0.0836,
+      "eval_runtime": 3345.1029,
+      "eval_samples_per_second": 0.272,
+      "eval_steps_per_second": 0.272,
       "step": 5250
     },
     {
-      "epoch": 5.02,
+      "epoch": 5.82,
       "learning_rate": 2.7647058823529416e-05,
-      "loss": 2.2957,
+      "loss": 2.2821,
       "step": 5300
     },
     {
-      "epoch": 5.07,
+      "epoch": 5.88,
       "learning_rate": 2.7352941176470593e-05,
-      "loss": 2.2123,
+      "loss": 2.2773,
       "step": 5350
     },
     {
-      "epoch": 5.12,
+      "epoch": 5.93,
       "learning_rate": 2.7058823529411766e-05,
-      "loss": 2.1858,
+      "loss": 2.2261,
       "step": 5400
     },
     {
-      "epoch": 5.17,
+      "epoch": 5.99,
       "learning_rate": 2.6764705882352942e-05,
-      "loss": 2.225,
+      "loss": 2.233,
       "step": 5450
     },
     {
-      "epoch": 5.21,
+      "epoch": 6.04,
       "learning_rate": 2.647058823529412e-05,
-      "loss": 2.2766,
+      "loss": 2.0471,
       "step": 5500
     },
     {
-      "epoch": 5.26,
+      "epoch": 6.1,
       "learning_rate": 2.6176470588235295e-05,
-      "loss": 2.228,
+      "loss": 2.0109,
       "step": 5550
     },
     {
-      "epoch": 5.31,
+      "epoch": 6.15,
       "learning_rate": 2.5882352941176475e-05,
-      "loss": 2.2788,
+      "loss": 2.0577,
       "step": 5600
     },
     {
-      "epoch": 5.36,
+      "epoch": 6.21,
       "learning_rate": 2.5588235294117645e-05,
-      "loss": 2.1984,
+      "loss": 2.0739,
       "step": 5650
     },
     {
-      "epoch": 5.4,
+      "epoch": 6.26,
       "learning_rate": 2.5294117647058825e-05,
-      "loss": 2.2213,
+      "loss": 2.0485,
       "step": 5700
     },
     {
-      "epoch": 5.45,
+      "epoch": 6.32,
       "learning_rate": 2.5e-05,
-      "loss": 2.1958,
+      "loss": 2.0153,
       "step": 5750
     },
     {
-      "epoch": 5.5,
+      "epoch": 6.37,
       "learning_rate": 2.4705882352941178e-05,
-      "loss": 2.2127,
+      "loss": 2.0995,
       "step": 5800
     },
     {
-      "epoch": 5.54,
+      "epoch": 6.43,
       "learning_rate": 2.4411764705882354e-05,
-      "loss": 2.3364,
+      "loss": 2.0478,
       "step": 5850
     },
     {
-      "epoch": 5.59,
+      "epoch": 6.48,
       "learning_rate": 2.411764705882353e-05,
-      "loss": 2.2505,
+      "loss": 2.0485,
       "step": 5900
     },
     {
-      "epoch": 5.64,
+      "epoch": 6.54,
       "learning_rate": 2.3823529411764707e-05,
-      "loss": 2.2923,
+      "loss": 2.0596,
       "step": 5950
     },
     {
-      "epoch": 5.69,
+      "epoch": 6.59,
       "learning_rate": 2.3529411764705884e-05,
-      "loss": 2.2209,
+      "loss": 2.0481,
       "step": 6000
     },
     {
-      "epoch": 5.69,
-      "eval_loss": 3.315330743789673,
-      "eval_rouge2_fmeasure": 0.0756,
-      "eval_rouge2_precision": 0.0903,
-      "eval_rouge2_recall": 0.0852,
-      "eval_runtime": 3241.2877,
-      "eval_samples_per_second": 0.29,
-      "eval_steps_per_second": 0.29,
+      "epoch": 6.59,
+      "eval_loss": 3.4252965450286865,
+      "eval_rouge2_fmeasure": 0.0694,
+      "eval_rouge2_precision": 0.0811,
+      "eval_rouge2_recall": 0.0809,
+      "eval_runtime": 3157.6234,
+      "eval_samples_per_second": 0.288,
+      "eval_steps_per_second": 0.288,
       "step": 6000
     },
     {
-      "epoch": 5.73,
+      "epoch": 6.65,
       "learning_rate": 2.323529411764706e-05,
-      "loss": 2.2958,
+      "loss": 2.0491,
       "step": 6050
     },
     {
-      "epoch": 5.78,
+      "epoch": 6.7,
       "learning_rate": 2.2941176470588237e-05,
-      "loss": 2.2355,
+      "loss": 2.1042,
       "step": 6100
     },
     {
-      "epoch": 5.83,
+      "epoch": 6.76,
       "learning_rate": 2.2647058823529413e-05,
-      "loss": 2.3148,
+      "loss": 2.0652,
       "step": 6150
     },
     {
-      "epoch": 5.88,
+      "epoch": 6.81,
       "learning_rate": 2.235294117647059e-05,
-      "loss": 2.2468,
+      "loss": 2.0637,
       "step": 6200
     },
     {
-      "epoch": 5.92,
+      "epoch": 6.87,
       "learning_rate": 2.2058823529411766e-05,
-      "loss": 2.2926,
+      "loss": 2.1071,
       "step": 6250
     },
     {
-      "epoch": 5.97,
+      "epoch": 6.92,
       "learning_rate": 2.1764705882352943e-05,
-      "loss": 2.2649,
+      "loss": 2.0651,
       "step": 6300
     },
     {
-      "epoch": 6.02,
+      "epoch": 6.98,
       "learning_rate": 2.1470588235294116e-05,
-      "loss": 2.2116,
+      "loss": 2.0646,
       "step": 6350
     },
     {
-      "epoch": 6.07,
+      "epoch": 7.03,
       "learning_rate": 2.1176470588235296e-05,
-      "loss": 2.0279,
+      "loss": 2.0564,
       "step": 6400
     },
     {
-      "epoch": 6.11,
+      "epoch": 7.09,
       "learning_rate": 2.0882352941176472e-05,
-      "loss": 2.1057,
+      "loss": 1.8977,
       "step": 6450
     },
     {
-      "epoch": 6.16,
+      "epoch": 7.14,
       "learning_rate": 2.058823529411765e-05,
-      "loss": 2.0691,
+      "loss": 1.8789,
       "step": 6500
     },
     {
-      "epoch": 6.21,
+      "epoch": 7.2,
       "learning_rate": 2.0294117647058825e-05,
-      "loss": 2.0331,
+      "loss": 1.9092,
       "step": 6550
     },
     {
-      "epoch": 6.26,
+      "epoch": 7.25,
       "learning_rate": 2e-05,
-      "loss": 2.1501,
+      "loss": 1.8689,
       "step": 6600
     },
     {
-      "epoch": 6.3,
+      "epoch": 7.31,
       "learning_rate": 1.9705882352941178e-05,
-      "loss": 2.0947,
+      "loss": 1.9511,
       "step": 6650
     },
     {
-      "epoch": 6.35,
+      "epoch": 7.36,
       "learning_rate": 1.9411764705882355e-05,
-      "loss": 2.1201,
+      "loss": 1.895,
       "step": 6700
     },
     {
-      "epoch": 6.4,
+      "epoch": 7.42,
       "learning_rate": 1.9117647058823528e-05,
-      "loss": 2.0768,
+      "loss": 1.938,
       "step": 6750
     },
     {
-      "epoch": 6.4,
-      "eval_loss": 3.3413684368133545,
-      "eval_rouge2_fmeasure": 0.0733,
-      "eval_rouge2_precision": 0.086,
-      "eval_rouge2_recall": 0.0847,
-      "eval_runtime": 3440.5432,
-      "eval_samples_per_second": 0.273,
-      "eval_steps_per_second": 0.273,
+      "epoch": 7.42,
+      "eval_loss": 3.5046603679656982,
+      "eval_rouge2_fmeasure": 0.0706,
+      "eval_rouge2_precision": 0.0808,
+      "eval_rouge2_recall": 0.0825,
+      "eval_runtime": 3282.9811,
+      "eval_samples_per_second": 0.277,
+      "eval_steps_per_second": 0.277,
       "step": 6750
     },
     {
-      "epoch": 6.45,
+      "epoch": 7.47,
       "learning_rate": 1.8823529411764708e-05,
-      "loss": 2.0365,
+      "loss": 1.9673,
       "step": 6800
     },
     {
-      "epoch": 6.49,
+      "epoch": 7.53,
       "learning_rate": 1.8529411764705884e-05,
-      "loss": 2.0608,
+      "loss": 1.9022,
       "step": 6850
     },
     {
-      "epoch": 6.54,
+      "epoch": 7.58,
       "learning_rate": 1.8235294117647057e-05,
-      "loss": 2.0621,
+      "loss": 1.9619,
       "step": 6900
     },
     {
-      "epoch": 6.59,
+      "epoch": 7.64,
       "learning_rate": 1.7941176470588237e-05,
-      "loss": 2.0557,
+      "loss": 1.9199,
       "step": 6950
     },
     {
-      "epoch": 6.63,
+      "epoch": 7.69,
       "learning_rate": 1.7647058823529414e-05,
-      "loss": 2.1465,
+      "loss": 1.9814,
       "step": 7000
     },
     {
-      "epoch": 6.68,
+      "epoch": 7.75,
       "learning_rate": 1.735294117647059e-05,
-      "loss": 2.1462,
+      "loss": 1.8905,
       "step": 7050
     },
     {
-      "epoch": 6.73,
+      "epoch": 7.8,
       "learning_rate": 1.7058823529411767e-05,
-      "loss": 2.0611,
+      "loss": 1.8699,
       "step": 7100
     },
     {
-      "epoch": 6.78,
+      "epoch": 7.86,
       "learning_rate": 1.676470588235294e-05,
-      "loss": 2.1656,
+      "loss": 1.8457,
       "step": 7150
     },
     {
-      "epoch": 6.82,
+      "epoch": 7.91,
       "learning_rate": 1.647058823529412e-05,
-      "loss": 2.1023,
+      "loss": 1.9926,
       "step": 7200
     },
     {
-      "epoch": 6.87,
+      "epoch": 7.97,
       "learning_rate": 1.6176470588235296e-05,
-      "loss": 2.1231,
+      "loss": 1.9444,
       "step": 7250
     },
     {
-      "epoch": 6.92,
+      "epoch": 8.02,
       "learning_rate": 1.588235294117647e-05,
-      "loss": 2.1836,
+      "loss": 1.9152,
       "step": 7300
     },
     {
-      "epoch": 6.97,
+      "epoch": 8.08,
       "learning_rate": 1.558823529411765e-05,
-      "loss": 2.1186,
+      "loss": 1.7481,
       "step": 7350
     },
     {
-      "epoch": 7.01,
+      "epoch": 8.13,
       "learning_rate": 1.5294117647058826e-05,
-      "loss": 2.0946,
+      "loss": 1.7448,
       "step": 7400
     },
     {
-      "epoch": 7.06,
+      "epoch": 8.19,
       "learning_rate": 1.5e-05,
-      "loss": 1.9738,
+      "loss": 1.7673,
       "step": 7450
     },
     {
-      "epoch": 7.11,
+      "epoch": 8.24,
       "learning_rate": 1.4705882352941177e-05,
-      "loss": 2.0296,
+      "loss": 1.8317,
       "step": 7500
     },
     {
-      "epoch": 7.11,
-      "eval_loss": 3.387500286102295,
-      "eval_rouge2_fmeasure": 0.0751,
-      "eval_rouge2_precision": 0.0879,
-      "eval_rouge2_recall": 0.086,
-      "eval_runtime": 3327.6543,
-      "eval_samples_per_second": 0.282,
-      "eval_steps_per_second": 0.282,
+      "epoch": 8.24,
+      "eval_loss": 3.5628507137298584,
+      "eval_rouge2_fmeasure": 0.0681,
+      "eval_rouge2_precision": 0.0793,
+      "eval_rouge2_recall": 0.0794,
+      "eval_runtime": 3215.1275,
+      "eval_samples_per_second": 0.283,
+      "eval_steps_per_second": 0.283,
       "step": 7500
     },
     {
-      "epoch": 7.16,
+      "epoch": 8.3,
       "learning_rate": 1.4411764705882352e-05,
-      "loss": 1.9327,
+      "loss": 1.7672,
       "step": 7550
     },
     {
-      "epoch": 7.2,
+      "epoch": 8.35,
       "learning_rate": 1.411764705882353e-05,
-      "loss": 1.967,
+      "loss": 1.7958,
       "step": 7600
     },
     {
-      "epoch": 7.25,
+      "epoch": 8.41,
       "learning_rate": 1.3823529411764708e-05,
-      "loss": 1.9642,
+      "loss": 1.7916,
       "step": 7650
     },
     {
-      "epoch": 7.3,
+      "epoch": 8.46,
       "learning_rate": 1.3529411764705883e-05,
-      "loss": 1.9403,
+      "loss": 1.783,
       "step": 7700
     },
     {
-      "epoch": 7.35,
+      "epoch": 8.52,
       "learning_rate": 1.323529411764706e-05,
-      "loss": 1.9594,
+      "loss": 1.8509,
       "step": 7750
     },
     {
-      "epoch": 7.39,
+      "epoch": 8.57,
       "learning_rate": 1.2941176470588238e-05,
-      "loss": 1.9846,
+      "loss": 1.8343,
       "step": 7800
     },
     {
-      "epoch": 7.44,
+      "epoch": 8.63,
       "learning_rate": 1.2647058823529412e-05,
-      "loss": 1.9939,
+      "loss": 1.827,
       "step": 7850
     },
     {
-      "epoch": 7.49,
+      "epoch": 8.68,
       "learning_rate": 1.2352941176470589e-05,
-      "loss": 1.9504,
+      "loss": 1.7483,
       "step": 7900
     },
     {
-      "epoch": 7.54,
+      "epoch": 8.74,
       "learning_rate": 1.2058823529411765e-05,
-      "loss": 2.0966,
+      "loss": 1.867,
       "step": 7950
     },
     {
-      "epoch": 7.58,
+      "epoch": 8.79,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 1.9842,
+      "loss": 1.9009,
       "step": 8000
     },
     {
-      "epoch": 7.63,
+      "epoch": 8.85,
       "learning_rate": 1.1470588235294118e-05,
-      "loss": 1.9927,
+      "loss": 1.8695,
       "step": 8050
     },
     {
-      "epoch": 7.68,
+      "epoch": 8.9,
       "learning_rate": 1.1176470588235295e-05,
-      "loss": 1.983,
+      "loss": 1.8261,
       "step": 8100
     },
     {
-      "epoch": 7.72,
+      "epoch": 8.96,
       "learning_rate": 1.0882352941176471e-05,
-      "loss": 1.9447,
+      "loss": 1.8723,
       "step": 8150
     },
     {
-      "epoch": 7.77,
+      "epoch": 9.01,
       "learning_rate": 1.0588235294117648e-05,
-      "loss": 1.9926,
+      "loss": 1.8381,
       "step": 8200
     },
     {
-      "epoch": 7.82,
+      "epoch": 9.07,
       "learning_rate": 1.0294117647058824e-05,
-      "loss": 1.9951,
+      "loss": 1.6888,
       "step": 8250
     },
     {
-      "epoch": 7.82,
-      "eval_loss": 3.400300979614258,
-      "eval_rouge2_fmeasure": 0.0763,
-      "eval_rouge2_precision": 0.0889,
-      "eval_rouge2_recall": 0.0874,
-      "eval_runtime": 3375.9502,
-      "eval_samples_per_second": 0.278,
-      "eval_steps_per_second": 0.278,
+      "epoch": 9.07,
+      "eval_loss": 3.601710557937622,
+      "eval_rouge2_fmeasure": 0.0702,
+      "eval_rouge2_precision": 0.0802,
+      "eval_rouge2_recall": 0.0823,
+      "eval_runtime": 3366.9405,
+      "eval_samples_per_second": 0.27,
+      "eval_steps_per_second": 0.27,
       "step": 8250
     },
     {
-      "epoch": 7.87,
+      "epoch": 9.12,
       "learning_rate": 1e-05,
-      "loss": 1.9629,
+      "loss": 1.6633,
       "step": 8300
     },
     {
-      "epoch": 7.91,
+      "epoch": 9.18,
       "learning_rate": 9.705882352941177e-06,
-      "loss": 2.0047,
+      "loss": 1.6752,
       "step": 8350
     },
     {
-      "epoch": 7.96,
+      "epoch": 9.23,
       "learning_rate": 9.411764705882354e-06,
-      "loss": 1.9716,
+      "loss": 1.7479,
       "step": 8400
     },
     {
-      "epoch": 8.01,
+      "epoch": 9.29,
       "learning_rate": 9.117647058823529e-06,
-      "loss": 1.936,
+      "loss": 1.8026,
       "step": 8450
     },
     {
-      "epoch": 8.06,
+      "epoch": 9.34,
       "learning_rate": 8.823529411764707e-06,
-      "loss": 1.8096,
+      "loss": 1.656,
       "step": 8500
     },
     {
-      "epoch": 8.1,
+      "epoch": 9.4,
       "learning_rate": 8.529411764705883e-06,
-      "loss": 1.93,
+      "loss": 1.7601,
       "step": 8550
     },
     {
-      "epoch": 8.15,
+      "epoch": 9.45,
       "learning_rate": 8.23529411764706e-06,
-      "loss": 1.8487,
+      "loss": 1.705,
       "step": 8600
     },
     {
-      "epoch": 8.2,
+      "epoch": 9.51,
       "learning_rate": 7.941176470588235e-06,
-      "loss": 1.8646,
+      "loss": 1.7438,
       "step": 8650
     },
     {
-      "epoch": 8.25,
+      "epoch": 9.56,
       "learning_rate": 7.647058823529413e-06,
-      "loss": 1.9603,
+      "loss": 1.7386,
       "step": 8700
     },
     {
-      "epoch": 8.29,
+      "epoch": 9.62,
       "learning_rate": 7.3529411764705884e-06,
-      "loss": 1.9469,
+      "loss": 1.6246,
       "step": 8750
     },
     {
-      "epoch": 8.34,
+      "epoch": 9.67,
       "learning_rate": 7.058823529411765e-06,
-      "loss": 1.8723,
+      "loss": 1.7827,
       "step": 8800
     },
     {
-      "epoch": 8.39,
+      "epoch": 9.73,
       "learning_rate": 6.7647058823529414e-06,
-      "loss": 1.9527,
+      "loss": 1.7536,
       "step": 8850
     },
     {
-      "epoch": 8.44,
+      "epoch": 9.78,
       "learning_rate": 6.470588235294119e-06,
-      "loss": 1.8656,
+      "loss": 1.754,
       "step": 8900
     },
     {
-      "epoch": 8.48,
+      "epoch": 9.83,
       "learning_rate": 6.1764705882352944e-06,
-      "loss": 1.9241,
+      "loss": 1.7615,
       "step": 8950
     },
     {
-      "epoch": 8.53,
+      "epoch": 9.89,
       "learning_rate": 5.882352941176471e-06,
-      "loss": 1.9125,
+      "loss": 1.8552,
       "step": 9000
     },
     {
-      "epoch": 8.53,
-      "eval_loss": 3.398467779159546,
-      "eval_rouge2_fmeasure": 0.0755,
-      "eval_rouge2_precision": 0.0875,
-      "eval_rouge2_recall": 0.0879,
-      "eval_runtime": 3459.2495,
-      "eval_samples_per_second": 0.271,
-      "eval_steps_per_second": 0.271,
+      "epoch": 9.89,
+      "eval_loss": 3.56019926071167,
+      "eval_rouge2_fmeasure": 0.0704,
+      "eval_rouge2_precision": 0.0804,
+      "eval_rouge2_recall": 0.0827,
+      "eval_runtime": 3403.6559,
+      "eval_samples_per_second": 0.267,
+      "eval_steps_per_second": 0.267,
       "step": 9000
     },
     {
-      "epoch": 8.58,
+      "epoch": 9.94,
       "learning_rate": 5.588235294117647e-06,
-      "loss": 1.9615,
+      "loss": 1.7489,
       "step": 9050
     },
     {
-      "epoch": 8.63,
+      "epoch": 10.0,
       "learning_rate": 5.294117647058824e-06,
-      "loss": 1.9059,
+      "loss": 1.6774,
       "step": 9100
     },
     {
-      "epoch": 8.67,
+      "epoch": 10.05,
       "learning_rate": 5e-06,
-      "loss": 1.9107,
+      "loss": 1.6038,
       "step": 9150
     },
     {
-      "epoch": 8.72,
+      "epoch": 10.11,
       "learning_rate": 4.705882352941177e-06,
-      "loss": 1.8266,
+      "loss": 1.6154,
       "step": 9200
     },
     {
-      "epoch": 8.77,
+      "epoch": 10.16,
       "learning_rate": 4.411764705882353e-06,
-      "loss": 1.7917,
+      "loss": 1.5998,
       "step": 9250
     },
     {
-      "epoch": 8.81,
+      "epoch": 10.22,
       "learning_rate": 4.11764705882353e-06,
-      "loss": 1.8893,
+      "loss": 1.6834,
       "step": 9300
     },
     {
-      "epoch": 8.86,
+      "epoch": 10.27,
       "learning_rate": 3.823529411764706e-06,
-      "loss": 1.9275,
+      "loss": 1.6449,
       "step": 9350
     },
     {
-      "epoch": 8.91,
+      "epoch": 10.33,
       "learning_rate": 3.5294117647058825e-06,
-      "loss": 1.8471,
+      "loss": 1.657,
       "step": 9400
     },
     {
-      "epoch": 8.96,
+      "epoch": 10.38,
       "learning_rate": 3.2352941176470594e-06,
-      "loss": 1.8878,
+      "loss": 1.7073,
       "step": 9450
     },
     {
-      "epoch": 9.0,
+      "epoch": 10.44,
       "learning_rate": 2.9411764705882355e-06,
-      "loss": 1.8799,
+      "loss": 1.7346,
       "step": 9500
     },
     {
-      "epoch": 9.05,
+      "epoch": 10.49,
       "learning_rate": 2.647058823529412e-06,
-      "loss": 1.919,
+      "loss": 1.6624,
       "step": 9550
     },
     {
-      "epoch": 9.1,
+      "epoch": 10.55,
       "learning_rate": 2.3529411764705885e-06,
-      "loss": 1.9169,
+      "loss": 1.6514,
       "step": 9600
     },
     {
-      "epoch": 9.15,
+      "epoch": 10.6,
       "learning_rate": 2.058823529411765e-06,
-      "loss": 1.7915,
+      "loss": 1.7821,
       "step": 9650
     },
     {
-      "epoch": 9.19,
+      "epoch": 10.66,
       "learning_rate": 1.7647058823529412e-06,
-      "loss": 1.755,
+      "loss": 1.7201,
       "step": 9700
     },
     {
-      "epoch": 9.24,
+      "epoch": 10.71,
       "learning_rate": 1.4705882352941177e-06,
-      "loss": 1.8538,
+      "loss": 1.7268,
       "step": 9750
     },
     {
-      "epoch": 9.24,
-      "eval_loss": 3.432481288909912,
-      "eval_rouge2_fmeasure": 0.0765,
-      "eval_rouge2_precision": 0.09,
-      "eval_rouge2_recall": 0.0879,
-      "eval_runtime": 3317.1381,
-      "eval_samples_per_second": 0.283,
-      "eval_steps_per_second": 0.283,
+      "epoch": 10.71,
+      "eval_loss": 3.5929648876190186,
+      "eval_rouge2_fmeasure": 0.0706,
+      "eval_rouge2_precision": 0.0806,
+      "eval_rouge2_recall": 0.0831,
+      "eval_runtime": 3342.0135,
+      "eval_samples_per_second": 0.272,
+      "eval_steps_per_second": 0.272,
       "step": 9750
     },
     {
-      "epoch": 9.29,
+      "epoch": 10.77,
       "learning_rate": 1.1764705882352942e-06,
-      "loss": 1.8775,
+      "loss": 1.6402,
       "step": 9800
     },
     {
-      "epoch": 9.34,
+      "epoch": 10.82,
       "learning_rate": 8.823529411764706e-07,
-      "loss": 1.8207,
+      "loss": 1.7222,
       "step": 9850
     },
     {
-      "epoch": 9.38,
+      "epoch": 10.88,
       "learning_rate": 5.882352941176471e-07,
-      "loss": 1.7702,
+      "loss": 1.605,
       "step": 9900
     },
     {
-      "epoch": 9.43,
+      "epoch": 10.93,
       "learning_rate": 2.9411764705882356e-07,
-      "loss": 1.8477,
+      "loss": 1.6892,
       "step": 9950
     },
     {
-      "epoch": 9.48,
+      "epoch": 10.99,
       "learning_rate": 0.0,
-      "loss": 1.8267,
+      "loss": 1.666,
       "step": 10000
     }
   ],
   "max_steps": 10000,
-  "num_train_epochs": 10,
-  "total_flos": 8.642599618892268e+17,
+  "num_train_epochs": 11,
+  "total_flos": 8.642815635279053e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2cc9948b5504197135c9464931aba1b542a873784597affbfc5c1a474cd95a37
+oid sha256:e0a3e73b30535a7e70fb03cfad6762a033bde029dd1ae3221f14c599ca9d401e
 size 3707