Upload 12 files

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc0721c949ea8c0652575a7a7965236d8430d208950c59b2512d24ef8e5a39e9
+oid sha256:b1159d5565655258033531e98b7e2d6c1cc2fb6ff46bbded5a46b93fe5b03697
 size 2178566661

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1ef6da9222f0db7ccec9a7714a6b0fa2b15788cf63bffde8e65de3af073cc50
+oid sha256:56860ad54851e9df34d10dd641120c6037d6d1fa7278b74c8006bce3f84ed66d
 size 1089301733

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1b9d7043ccd8afec2aa413640303022422f0689c0ef3679c13bc1ab6ad7b200
+oid sha256:d5a9fe30454160ce0bf0de298e75663580247deda90afc4dad314abf874d5c8b
 size 14575

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37565a3fc8b8a3d73ae2398cca6bd61a01fe9813fd4bb04d4f2cd5b97983d762
+oid sha256:286f87400850907e18b85d6b5636d21dddc64081bed824506f1291159e62fba6
 size 627

tokenizer.json

@@ -1,21 +1,7 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
-  "padding": {
-    "strategy": {
-      "Fixed": 512
-    },
-    "direction": "Right",
-    "pad_to_multiple_of": null,
-    "pad_id": 0,
-    "pad_type_id": 0,
-    "pad_token": "<pad>"
-  },
+  "truncation": null,
+  "padding": null,
   "added_tokens": [
     {
       "id": 0,

trainer_state.json

@@ -1,318 +1,399 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 7.582800284292822,
-  "global_step": 8000,
+  "epoch": 10.988739357319417,
+  "global_step": 10000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.24,
+      "epoch": 0.27,
       "learning_rate": 8.333333333333334e-06,
-      "loss": 5.2679,
+      "loss": 5.2801,
       "step": 250
     },
     {
-      "epoch": 0.47,
+      "epoch": 0.55,
       "learning_rate": 1.6666666666666667e-05,
-      "loss": 4.0457,
+      "loss": 4.0432,
       "step": 500
     },
     {
-      "epoch": 0.71,
+      "epoch": 0.82,
       "learning_rate": 2.5e-05,
-      "loss": 3.7746,
+      "loss": 3.7257,
       "step": 750
     },
     {
-      "epoch": 0.71,
-      "eval_loss": 3.4735970497131348,
-      "eval_rouge2_fmeasure": 0.0608,
-      "eval_rouge2_precision": 0.0752,
-      "eval_rouge2_recall": 0.0676,
-      "eval_runtime": 10444.2518,
-      "eval_samples_per_second": 0.09,
-      "eval_steps_per_second": 0.09,
+      "epoch": 0.82,
+      "eval_loss": 3.4456169605255127,
+      "eval_rouge2_fmeasure": 0.0619,
+      "eval_rouge2_precision": 0.0751,
+      "eval_rouge2_recall": 0.0686,
+      "eval_runtime": 10201.4024,
+      "eval_samples_per_second": 0.089,
+      "eval_steps_per_second": 0.089,
       "step": 750
     },
     {
-      "epoch": 0.95,
+      "epoch": 1.1,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 3.6193,
+      "loss": 3.6021,
       "step": 1000
     },
     {
-      "epoch": 1.18,
+      "epoch": 1.37,
       "learning_rate": 4.166666666666667e-05,
-      "loss": 3.5271,
+      "loss": 3.505,
       "step": 1250
     },
     {
-      "epoch": 1.42,
+      "epoch": 1.65,
       "learning_rate": 5e-05,
-      "loss": 3.4793,
+      "loss": 3.4611,
       "step": 1500
     },
     {
-      "epoch": 1.42,
-      "eval_loss": 3.331468105316162,
-      "eval_rouge2_fmeasure": 0.0653,
-      "eval_rouge2_precision": 0.0796,
-      "eval_rouge2_recall": 0.0735,
-      "eval_runtime": 10901.7035,
-      "eval_samples_per_second": 0.086,
-      "eval_steps_per_second": 0.086,
+      "epoch": 1.65,
+      "eval_loss": 3.324451446533203,
+      "eval_rouge2_fmeasure": 0.064,
+      "eval_rouge2_precision": 0.0771,
+      "eval_rouge2_recall": 0.0724,
+      "eval_runtime": 10238.3889,
+      "eval_samples_per_second": 0.089,
+      "eval_steps_per_second": 0.089,
       "step": 1500
     },
     {
-      "epoch": 1.66,
+      "epoch": 1.92,
       "learning_rate": 4.8529411764705885e-05,
-      "loss": 3.4251,
+      "loss": 3.4327,
       "step": 1750
     },
     {
-      "epoch": 1.9,
+      "epoch": 2.2,
       "learning_rate": 4.705882352941177e-05,
-      "loss": 3.4313,
+      "loss": 3.3307,
       "step": 2000
     },
     {
-      "epoch": 2.13,
+      "epoch": 2.47,
       "learning_rate": 4.558823529411765e-05,
-      "loss": 3.3459,
+      "loss": 3.2866,
       "step": 2250
     },
     {
-      "epoch": 2.13,
-      "eval_loss": 3.2606232166290283,
-      "eval_rouge2_fmeasure": 0.0685,
-      "eval_rouge2_precision": 0.0825,
-      "eval_rouge2_recall": 0.077,
-      "eval_runtime": 10929.9339,
-      "eval_samples_per_second": 0.086,
-      "eval_steps_per_second": 0.086,
+      "epoch": 2.47,
+      "eval_loss": 3.268772602081299,
+      "eval_rouge2_fmeasure": 0.0675,
+      "eval_rouge2_precision": 0.0787,
+      "eval_rouge2_recall": 0.0787,
+      "eval_runtime": 11096.9726,
+      "eval_samples_per_second": 0.082,
+      "eval_steps_per_second": 0.082,
       "step": 2250
     },
     {
-      "epoch": 2.37,
+      "epoch": 2.75,
       "learning_rate": 4.411764705882353e-05,
-      "loss": 3.291,
+      "loss": 3.3072,
       "step": 2500
     },
     {
-      "epoch": 2.61,
+      "epoch": 3.02,
       "learning_rate": 4.2647058823529415e-05,
-      "loss": 3.2803,
+      "loss": 3.2516,
       "step": 2750
     },
     {
-      "epoch": 2.84,
+      "epoch": 3.3,
       "learning_rate": 4.11764705882353e-05,
-      "loss": 3.2948,
+      "loss": 3.1896,
       "step": 3000
     },
     {
-      "epoch": 2.84,
-      "eval_loss": 3.227094888687134,
-      "eval_rouge2_fmeasure": 0.0684,
-      "eval_rouge2_precision": 0.0815,
-      "eval_rouge2_recall": 0.078,
-      "eval_runtime": 11536.812,
-      "eval_samples_per_second": 0.081,
-      "eval_steps_per_second": 0.081,
+      "epoch": 3.3,
+      "eval_loss": 3.2407822608947754,
+      "eval_rouge2_fmeasure": 0.0688,
+      "eval_rouge2_precision": 0.0806,
+      "eval_rouge2_recall": 0.0802,
+      "eval_runtime": 10753.114,
+      "eval_samples_per_second": 0.085,
+      "eval_steps_per_second": 0.085,
       "step": 3000
     },
     {
-      "epoch": 3.08,
+      "epoch": 3.57,
       "learning_rate": 3.970588235294117e-05,
-      "loss": 3.2314,
+      "loss": 3.1495,
       "step": 3250
     },
     {
-      "epoch": 3.32,
+      "epoch": 3.85,
       "learning_rate": 3.8235294117647055e-05,
-      "loss": 3.2134,
+      "loss": 3.1904,
       "step": 3500
     },
     {
-      "epoch": 3.55,
+      "epoch": 4.12,
       "learning_rate": 3.6764705882352945e-05,
-      "loss": 3.1879,
+      "loss": 3.1541,
       "step": 3750
     },
     {
-      "epoch": 3.55,
-      "eval_loss": 3.2067530155181885,
+      "epoch": 4.12,
+      "eval_loss": 3.230734348297119,
       "eval_rouge2_fmeasure": 0.0704,
-      "eval_rouge2_precision": 0.083,
-      "eval_rouge2_recall": 0.0809,
-      "eval_runtime": 11358.1561,
-      "eval_samples_per_second": 0.083,
-      "eval_steps_per_second": 0.083,
+      "eval_rouge2_precision": 0.081,
+      "eval_rouge2_recall": 0.083,
+      "eval_runtime": 11594.4275,
+      "eval_samples_per_second": 0.078,
+      "eval_steps_per_second": 0.078,
       "step": 3750
     },
     {
-      "epoch": 3.79,
+      "epoch": 4.4,
       "learning_rate": 3.529411764705883e-05,
-      "loss": 3.1617,
+      "loss": 3.0879,
       "step": 4000
     },
     {
-      "epoch": 4.03,
+      "epoch": 4.67,
       "learning_rate": 3.382352941176471e-05,
-      "loss": 3.172,
+      "loss": 3.0753,
       "step": 4250
     },
     {
-      "epoch": 4.27,
+      "epoch": 4.94,
       "learning_rate": 3.235294117647059e-05,
-      "loss": 3.1038,
+      "loss": 3.1243,
       "step": 4500
     },
     {
-      "epoch": 4.27,
-      "eval_loss": 3.1938178539276123,
-      "eval_rouge2_fmeasure": 0.0716,
-      "eval_rouge2_precision": 0.0858,
-      "eval_rouge2_recall": 0.0813,
-      "eval_runtime": 10958.7861,
-      "eval_samples_per_second": 0.086,
-      "eval_steps_per_second": 0.086,
+      "epoch": 4.94,
+      "eval_loss": 3.2160983085632324,
+      "eval_rouge2_fmeasure": 0.0707,
+      "eval_rouge2_precision": 0.0821,
+      "eval_rouge2_recall": 0.0817,
+      "eval_runtime": 11304.2042,
+      "eval_samples_per_second": 0.081,
+      "eval_steps_per_second": 0.081,
       "step": 4500
     },
     {
-      "epoch": 4.5,
+      "epoch": 5.22,
       "learning_rate": 3.0882352941176475e-05,
-      "loss": 3.1153,
+      "loss": 3.0364,
       "step": 4750
     },
     {
-      "epoch": 4.74,
+      "epoch": 5.49,
       "learning_rate": 2.9411764705882354e-05,
-      "loss": 3.1037,
+      "loss": 3.0249,
       "step": 5000
     },
     {
-      "epoch": 4.98,
+      "epoch": 5.77,
       "learning_rate": 2.7941176470588236e-05,
-      "loss": 3.1155,
+      "loss": 3.0428,
       "step": 5250
     },
     {
-      "epoch": 4.98,
-      "eval_loss": 3.187068462371826,
-      "eval_rouge2_fmeasure": 0.0725,
-      "eval_rouge2_precision": 0.0856,
-      "eval_rouge2_recall": 0.0839,
-      "eval_runtime": 11370.9481,
-      "eval_samples_per_second": 0.083,
-      "eval_steps_per_second": 0.083,
+      "epoch": 5.77,
+      "eval_loss": 3.2113523483276367,
+      "eval_rouge2_fmeasure": 0.072,
+      "eval_rouge2_precision": 0.0828,
+      "eval_rouge2_recall": 0.0848,
+      "eval_runtime": 11113.1913,
+      "eval_samples_per_second": 0.082,
+      "eval_steps_per_second": 0.082,
       "step": 5250
     },
     {
-      "epoch": 5.21,
+      "epoch": 6.04,
       "learning_rate": 2.647058823529412e-05,
-      "loss": 3.048,
+      "loss": 3.0229,
       "step": 5500
     },
     {
-      "epoch": 5.45,
+      "epoch": 6.32,
       "learning_rate": 2.5e-05,
-      "loss": 3.0603,
+      "loss": 2.979,
       "step": 5750
     },
     {
-      "epoch": 5.69,
+      "epoch": 6.59,
       "learning_rate": 2.3529411764705884e-05,
-      "loss": 3.0534,
+      "loss": 2.975,
       "step": 6000
     },
     {
-      "epoch": 5.69,
-      "eval_loss": 3.181117057800293,
-      "eval_rouge2_fmeasure": 0.0735,
-      "eval_rouge2_precision": 0.0865,
-      "eval_rouge2_recall": 0.0848,
-      "eval_runtime": 11629.6122,
-      "eval_samples_per_second": 0.081,
-      "eval_steps_per_second": 0.081,
+      "epoch": 6.59,
+      "eval_loss": 3.213066339492798,
+      "eval_rouge2_fmeasure": 0.0728,
+      "eval_rouge2_precision": 0.0841,
+      "eval_rouge2_recall": 0.085,
+      "eval_runtime": 11135.6948,
+      "eval_samples_per_second": 0.082,
+      "eval_steps_per_second": 0.082,
       "step": 6000
     },
     {
-      "epoch": 5.92,
+      "epoch": 6.87,
       "learning_rate": 2.2058823529411766e-05,
-      "loss": 3.0664,
+      "loss": 3.0022,
       "step": 6250
     },
     {
-      "epoch": 6.16,
+      "epoch": 7.14,
       "learning_rate": 2.058823529411765e-05,
-      "loss": 3.0255,
+      "loss": 2.9825,
       "step": 6500
     },
     {
-      "epoch": 6.4,
+      "epoch": 7.42,
       "learning_rate": 1.9117647058823528e-05,
-      "loss": 3.0338,
+      "loss": 2.9392,
       "step": 6750
     },
     {
-      "epoch": 6.4,
-      "eval_loss": 3.1781060695648193,
-      "eval_rouge2_fmeasure": 0.0747,
-      "eval_rouge2_precision": 0.0879,
-      "eval_rouge2_recall": 0.0865,
-      "eval_runtime": 11275.9298,
-      "eval_samples_per_second": 0.083,
-      "eval_steps_per_second": 0.083,
+      "epoch": 7.42,
+      "eval_loss": 3.2107818126678467,
+      "eval_rouge2_fmeasure": 0.073,
+      "eval_rouge2_precision": 0.0839,
+      "eval_rouge2_recall": 0.0848,
+      "eval_runtime": 11088.831,
+      "eval_samples_per_second": 0.082,
+      "eval_steps_per_second": 0.082,
       "step": 6750
     },
     {
-      "epoch": 6.63,
+      "epoch": 7.69,
       "learning_rate": 1.7647058823529414e-05,
-      "loss": 2.9872,
+      "loss": 2.9429,
       "step": 7000
     },
     {
-      "epoch": 6.87,
+      "epoch": 7.97,
       "learning_rate": 1.6176470588235296e-05,
-      "loss": 3.0128,
+      "loss": 2.9304,
       "step": 7250
     },
     {
-      "epoch": 7.11,
+      "epoch": 8.24,
       "learning_rate": 1.4705882352941177e-05,
-      "loss": 3.0266,
+      "loss": 2.9204,
       "step": 7500
     },
     {
-      "epoch": 7.11,
-      "eval_loss": 3.1731793880462646,
-      "eval_rouge2_fmeasure": 0.074,
-      "eval_rouge2_precision": 0.0886,
-      "eval_rouge2_recall": 0.0836,
-      "eval_runtime": 11197.1963,
-      "eval_samples_per_second": 0.084,
-      "eval_steps_per_second": 0.084,
+      "epoch": 8.24,
+      "eval_loss": 3.2088212966918945,
+      "eval_rouge2_fmeasure": 0.073,
+      "eval_rouge2_precision": 0.0833,
+      "eval_rouge2_recall": 0.0865,
+      "eval_runtime": 11327.9603,
+      "eval_samples_per_second": 0.08,
+      "eval_steps_per_second": 0.08,
       "step": 7500
     },
     {
-      "epoch": 7.35,
+      "epoch": 8.52,
       "learning_rate": 1.323529411764706e-05,
-      "loss": 2.9676,
+      "loss": 2.8969,
       "step": 7750
     },
     {
-      "epoch": 7.58,
+      "epoch": 8.79,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 2.9834,
+      "loss": 2.9162,
       "step": 8000
+    },
+    {
+      "epoch": 9.07,
+      "learning_rate": 1.0294117647058824e-05,
+      "loss": 2.9233,
+      "step": 8250
+    },
+    {
+      "epoch": 9.07,
+      "eval_loss": 3.210251808166504,
+      "eval_rouge2_fmeasure": 0.0738,
+      "eval_rouge2_precision": 0.0849,
+      "eval_rouge2_recall": 0.0859,
+      "eval_runtime": 11276.9504,
+      "eval_samples_per_second": 0.081,
+      "eval_steps_per_second": 0.081,
+      "step": 8250
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 8.823529411764707e-06,
+      "loss": 2.8888,
+      "step": 8500
+    },
+    {
+      "epoch": 9.62,
+      "learning_rate": 7.3529411764705884e-06,
+      "loss": 2.8771,
+      "step": 8750
+    },
+    {
+      "epoch": 9.89,
+      "learning_rate": 5.882352941176471e-06,
+      "loss": 2.9142,
+      "step": 9000
+    },
+    {
+      "epoch": 9.89,
+      "eval_loss": 3.206127643585205,
+      "eval_rouge2_fmeasure": 0.0728,
+      "eval_rouge2_precision": 0.0831,
+      "eval_rouge2_recall": 0.0855,
+      "eval_runtime": 11283.9811,
+      "eval_samples_per_second": 0.081,
+      "eval_steps_per_second": 0.081,
+      "step": 9000
+    },
+    {
+      "epoch": 10.16,
+      "learning_rate": 4.411764705882353e-06,
+      "loss": 2.857,
+      "step": 9250
+    },
+    {
+      "epoch": 10.44,
+      "learning_rate": 2.9411764705882355e-06,
+      "loss": 2.8846,
+      "step": 9500
+    },
+    {
+      "epoch": 10.71,
+      "learning_rate": 1.4705882352941177e-06,
+      "loss": 2.8889,
+      "step": 9750
+    },
+    {
+      "epoch": 10.71,
+      "eval_loss": 3.2091023921966553,
+      "eval_rouge2_fmeasure": 0.0729,
+      "eval_rouge2_precision": 0.0835,
+      "eval_rouge2_recall": 0.0849,
+      "eval_runtime": 11160.0848,
+      "eval_samples_per_second": 0.082,
+      "eval_steps_per_second": 0.082,
+      "step": 9750
+    },
+    {
+      "epoch": 10.99,
+      "learning_rate": 0.0,
+      "loss": 2.8707,
+      "step": 10000
     }
   ],
   "max_steps": 10000,
-  "num_train_epochs": 10,
-  "total_flos": 1.248414149407408e+18,
+  "num_train_epochs": 11,
+  "total_flos": 1.560566442271703e+18,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b1cbd06825f5f91c2f78f18190e09bad91e0918abcd1a637bcb25452f8f80e9
+oid sha256:a9986b0b87d9d3689e1844137c4bb44f4bd7b447d5fa7010df7b1ac978ea1e50
 size 3707