Upload 9 files

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ec2761edd5166d89d35a2049202efcb3d1b93545d235bc52ef77719be325d1a
+oid sha256:2b6196274b15ecaba8ff351fc9738fb49e7483a98ff2033e8c1bca2cb96a90f4
 size 501634165

train_results.txt

@@ -1,3 +1,3 @@
-epoch = 2.0
-train_runtime = 1295.503
-train_samples_per_second = 17.38
+epoch = 3.0
+train_runtime = 1945.6382
+train_samples_per_second = 17.359

trainer_state.json

@@ -1,292 +1,424 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.0,
-  "global_step": 22516,
+  "epoch": 3.0,
+  "global_step": 33774,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.04,
-      "learning_rate": 2.9333807070527625e-05,
-      "loss": 0.5737,
+      "learning_rate": 2.9555871380351753e-05,
+      "loss": 0.5739,
       "step": 500
     },
     {
       "epoch": 0.09,
-      "learning_rate": 2.8667614141055253e-05,
-      "loss": 0.4695,
+      "learning_rate": 2.91117427607035e-05,
+      "loss": 0.4701,
       "step": 1000
     },
     {
       "epoch": 0.13,
-      "learning_rate": 2.8001421211582874e-05,
-      "loss": 0.439,
+      "learning_rate": 2.8667614141055253e-05,
+      "loss": 0.4401,
       "step": 1500
     },
     {
       "epoch": 0.18,
-      "learning_rate": 2.73352282821105e-05,
-      "loss": 0.3985,
+      "learning_rate": 2.8223485521406998e-05,
+      "loss": 0.4007,
       "step": 2000
     },
     {
       "epoch": 0.22,
-      "learning_rate": 2.6669035352638126e-05,
-      "loss": 0.3858,
+      "learning_rate": 2.777935690175875e-05,
+      "loss": 0.3886,
       "step": 2500
     },
     {
       "epoch": 0.27,
-      "learning_rate": 2.600284242316575e-05,
-      "loss": 0.3637,
+      "learning_rate": 2.73352282821105e-05,
+      "loss": 0.3668,
       "step": 3000
     },
     {
       "epoch": 0.31,
-      "learning_rate": 2.5336649493693375e-05,
-      "loss": 0.3589,
+      "learning_rate": 2.689109966246225e-05,
+      "loss": 0.3622,
       "step": 3500
     },
     {
       "epoch": 0.36,
-      "learning_rate": 2.4670456564221e-05,
-      "loss": 0.3473,
+      "learning_rate": 2.6446971042814e-05,
+      "loss": 0.3513,
       "step": 4000
     },
     {
       "epoch": 0.4,
-      "learning_rate": 2.4004263634748624e-05,
-      "loss": 0.3185,
+      "learning_rate": 2.600284242316575e-05,
+      "loss": 0.3233,
       "step": 4500
     },
     {
       "epoch": 0.44,
-      "learning_rate": 2.3338070705276248e-05,
-      "loss": 0.3282,
+      "learning_rate": 2.55587138035175e-05,
+      "loss": 0.3341,
       "step": 5000
     },
     {
       "epoch": 0.49,
-      "learning_rate": 2.2671877775803876e-05,
-      "loss": 0.3195,
+      "learning_rate": 2.5114585183869247e-05,
+      "loss": 0.325,
       "step": 5500
     },
     {
       "epoch": 0.53,
-      "learning_rate": 2.20056848463315e-05,
-      "loss": 0.3099,
+      "learning_rate": 2.4670456564221e-05,
+      "loss": 0.3164,
       "step": 6000
     },
     {
       "epoch": 0.58,
-      "learning_rate": 2.133949191685912e-05,
-      "loss": 0.2884,
+      "learning_rate": 2.4226327944572748e-05,
+      "loss": 0.2953,
       "step": 6500
     },
     {
       "epoch": 0.62,
-      "learning_rate": 2.0673298987386746e-05,
-      "loss": 0.2962,
+      "learning_rate": 2.37821993249245e-05,
+      "loss": 0.3033,
       "step": 7000
     },
     {
       "epoch": 0.67,
-      "learning_rate": 2.0007106057914373e-05,
-      "loss": 0.2811,
+      "learning_rate": 2.3338070705276248e-05,
+      "loss": 0.2875,
       "step": 7500
     },
     {
       "epoch": 0.71,
-      "learning_rate": 1.9340913128441998e-05,
-      "loss": 0.2757,
+      "learning_rate": 2.2893942085628e-05,
+      "loss": 0.2848,
       "step": 8000
     },
     {
       "epoch": 0.76,
-      "learning_rate": 1.8674720198969622e-05,
-      "loss": 0.2615,
+      "learning_rate": 2.244981346597975e-05,
+      "loss": 0.27,
       "step": 8500
     },
     {
       "epoch": 0.8,
-      "learning_rate": 1.8008527269497247e-05,
-      "loss": 0.2716,
+      "learning_rate": 2.20056848463315e-05,
+      "loss": 0.2822,
       "step": 9000
     },
     {
       "epoch": 0.84,
-      "learning_rate": 1.734233434002487e-05,
-      "loss": 0.2503,
+      "learning_rate": 2.1561556226683245e-05,
+      "loss": 0.2595,
       "step": 9500
     },
     {
       "epoch": 0.89,
-      "learning_rate": 1.6676141410552495e-05,
-      "loss": 0.248,
+      "learning_rate": 2.1117427607034997e-05,
+      "loss": 0.2592,
       "step": 10000
     },
     {
       "epoch": 0.93,
-      "learning_rate": 1.6009948481080123e-05,
-      "loss": 0.2374,
+      "learning_rate": 2.0673298987386746e-05,
+      "loss": 0.2493,
       "step": 10500
     },
     {
       "epoch": 0.98,
-      "learning_rate": 1.5343755551607744e-05,
-      "loss": 0.247,
+      "learning_rate": 2.0229170367738498e-05,
+      "loss": 0.2574,
       "step": 11000
     },
     {
       "epoch": 1.02,
-      "learning_rate": 1.4677562622135372e-05,
-      "loss": 0.1944,
+      "learning_rate": 1.978504174809025e-05,
+      "loss": 0.2059,
       "step": 11500
     },
     {
       "epoch": 1.07,
-      "learning_rate": 1.4011369692662995e-05,
-      "loss": 0.1465,
+      "learning_rate": 1.9340913128441998e-05,
+      "loss": 0.1537,
       "step": 12000
     },
     {
       "epoch": 1.11,
-      "learning_rate": 1.334517676319062e-05,
-      "loss": 0.1489,
+      "learning_rate": 1.889678450879375e-05,
+      "loss": 0.1588,
       "step": 12500
     },
     {
       "epoch": 1.15,
-      "learning_rate": 1.2678983833718245e-05,
-      "loss": 0.1389,
+      "learning_rate": 1.8452655889145495e-05,
+      "loss": 0.1487,
       "step": 13000
     },
     {
       "epoch": 1.2,
-      "learning_rate": 1.201279090424587e-05,
-      "loss": 0.1495,
+      "learning_rate": 1.8008527269497247e-05,
+      "loss": 0.1608,
       "step": 13500
     },
     {
       "epoch": 1.24,
-      "learning_rate": 1.1346597974773496e-05,
-      "loss": 0.1483,
+      "learning_rate": 1.7564398649848995e-05,
+      "loss": 0.1619,
       "step": 14000
     },
     {
       "epoch": 1.29,
-      "learning_rate": 1.068040504530112e-05,
-      "loss": 0.1432,
+      "learning_rate": 1.7120270030200747e-05,
+      "loss": 0.155,
       "step": 14500
     },
     {
       "epoch": 1.33,
-      "learning_rate": 1.0014212115828744e-05,
-      "loss": 0.1288,
+      "learning_rate": 1.6676141410552495e-05,
+      "loss": 0.1418,
       "step": 15000
     },
     {
       "epoch": 1.38,
-      "learning_rate": 9.348019186356369e-06,
-      "loss": 0.1403,
+      "learning_rate": 1.6232012790904247e-05,
+      "loss": 0.1558,
       "step": 15500
     },
     {
       "epoch": 1.42,
-      "learning_rate": 8.681826256883993e-06,
-      "loss": 0.14,
+      "learning_rate": 1.5787884171255996e-05,
+      "loss": 0.155,
       "step": 16000
     },
     {
       "epoch": 1.47,
-      "learning_rate": 8.015633327411618e-06,
-      "loss": 0.1369,
+      "learning_rate": 1.5343755551607744e-05,
+      "loss": 0.1522,
       "step": 16500
     },
     {
       "epoch": 1.51,
-      "learning_rate": 7.349440397939243e-06,
-      "loss": 0.1241,
+      "learning_rate": 1.4899626931959496e-05,
+      "loss": 0.1386,
       "step": 17000
     },
     {
       "epoch": 1.55,
-      "learning_rate": 6.683247468466868e-06,
-      "loss": 0.129,
+      "learning_rate": 1.4455498312311245e-05,
+      "loss": 0.1442,
       "step": 17500
     },
     {
       "epoch": 1.6,
-      "learning_rate": 6.017054538994493e-06,
-      "loss": 0.1224,
+      "learning_rate": 1.4011369692662995e-05,
+      "loss": 0.1382,
       "step": 18000
     },
     {
       "epoch": 1.64,
-      "learning_rate": 5.350861609522118e-06,
-      "loss": 0.127,
+      "learning_rate": 1.3567241073014745e-05,
+      "loss": 0.1437,
       "step": 18500
     },
     {
       "epoch": 1.69,
-      "learning_rate": 4.684668680049743e-06,
-      "loss": 0.1258,
+      "learning_rate": 1.3123112453366497e-05,
+      "loss": 0.1405,
       "step": 19000
     },
     {
       "epoch": 1.73,
-      "learning_rate": 4.018475750577367e-06,
-      "loss": 0.1186,
+      "learning_rate": 1.2678983833718245e-05,
+      "loss": 0.1357,
       "step": 19500
     },
     {
       "epoch": 1.78,
-      "learning_rate": 3.3522828211049918e-06,
-      "loss": 0.1107,
+      "learning_rate": 1.2234855214069995e-05,
+      "loss": 0.125,
       "step": 20000
     },
     {
       "epoch": 1.82,
-      "learning_rate": 2.686089891632617e-06,
-      "loss": 0.1202,
+      "learning_rate": 1.1790726594421745e-05,
+      "loss": 0.1376,
       "step": 20500
     },
     {
       "epoch": 1.87,
-      "learning_rate": 2.0198969621602414e-06,
-      "loss": 0.1182,
+      "learning_rate": 1.1346597974773496e-05,
+      "loss": 0.136,
       "step": 21000
     },
     {
       "epoch": 1.91,
-      "learning_rate": 1.3537040326878664e-06,
-      "loss": 0.1132,
+      "learning_rate": 1.0902469355125244e-05,
+      "loss": 0.1251,
       "step": 21500
     },
     {
       "epoch": 1.95,
-      "learning_rate": 6.875111032154912e-07,
-      "loss": 0.112,
+      "learning_rate": 1.0458340735476994e-05,
+      "loss": 0.1258,
       "step": 22000
     },
     {
       "epoch": 2.0,
-      "learning_rate": 2.1318173743116005e-08,
-      "loss": 0.1181,
+      "learning_rate": 1.0014212115828744e-05,
+      "loss": 0.1303,
       "step": 22500
     },
     {
-      "epoch": 2.0,
-      "step": 22516,
-      "train_runtime": 1295.503,
-      "train_samples_per_second": 17.38
+      "epoch": 2.04,
+      "learning_rate": 9.570083496180493e-06,
+      "loss": 0.0792,
+      "step": 23000
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 9.125954876532243e-06,
+      "loss": 0.0794,
+      "step": 23500
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 8.681826256883993e-06,
+      "loss": 0.0781,
+      "step": 24000
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 8.237697637235745e-06,
+      "loss": 0.0777,
+      "step": 24500
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 7.793569017587494e-06,
+      "loss": 0.077,
+      "step": 25000
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 7.349440397939243e-06,
+      "loss": 0.0792,
+      "step": 25500
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 6.905311778290994e-06,
+      "loss": 0.0774,
+      "step": 26000
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 6.461183158642743e-06,
+      "loss": 0.0762,
+      "step": 26500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 6.017054538994493e-06,
+      "loss": 0.0744,
+      "step": 27000
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 5.572925919346243e-06,
+      "loss": 0.0725,
+      "step": 27500
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 5.128797299697992e-06,
+      "loss": 0.076,
+      "step": 28000
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 4.684668680049743e-06,
+      "loss": 0.0682,
+      "step": 28500
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 4.240540060401492e-06,
+      "loss": 0.074,
+      "step": 29000
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 3.7964114407532424e-06,
+      "loss": 0.0738,
+      "step": 29500
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 3.3522828211049918e-06,
+      "loss": 0.0678,
+      "step": 30000
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 2.908154201456742e-06,
+      "loss": 0.0668,
+      "step": 30500
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 2.4640255818084917e-06,
+      "loss": 0.0685,
+      "step": 31000
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.0198969621602414e-06,
+      "loss": 0.0668,
+      "step": 31500
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.5757683425119916e-06,
+      "loss": 0.068,
+      "step": 32000
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 1.1316397228637413e-06,
+      "loss": 0.0663,
+      "step": 32500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 6.875111032154912e-07,
+      "loss": 0.0629,
+      "step": 33000
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 2.4338248356724103e-07,
+      "loss": 0.0676,
+      "step": 33500
+    },
+    {
+      "epoch": 3.0,
+      "step": 33774,
+      "train_runtime": 1945.6382,
+      "train_samples_per_second": 17.359
     }
   ],
-  "max_steps": 22516,
-  "num_train_epochs": 2,
-  "total_flos": 34691312231055360,
+  "max_steps": 33774,
+  "num_train_epochs": 3,
+  "total_flos": 52036968346583040,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d39439ac6c3a4c50ff0a057635bb777618d30eaf62c7ac11198bfb012d8e4737
+oid sha256:5ea6da2311f8ad0e07c33c998a50bd07f1b7dba25e67f9bbe18dcb1cd7760aad
 size 2031