Training in progress, step 1000, checkpoint

checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:503a51c27e405557590c68fff2960e6708474dcd64e54d3a3be14c6f4dcf0480
+oid sha256:13e26e659feaf55dc24c2816fddddf43cdcd05cbeaeb5416150cbe3862cdc2f5
 size 997351674

checkpoint-1000/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4468695e1b2d761710940bf0cdf79d2ffb49023a4f5f8835e17e01806bf34823
+oid sha256:6f38b84adbe09d0344d1dbbbf1f09d4bf5ade258eda4b068c79156d5f89499a1
 size 498661166

checkpoint-1000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7619542dcc3c020675dd12b61cd559ace537a936c7aa3f24dcd8e91aa50b148
+oid sha256:8bcf00aea2ab4a621d89b6c23d69c7db0135ebd4afef60cb67dec55ce69fb934
 size 14244

checkpoint-1000/trainer_state.json

@@ -1,799 +1,1269 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.7524454477050414,
-  "eval_steps": 50,
+  "epoch": 3.0097817908201656,
+  "eval_steps": 20,
   "global_step": 1000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.01,
+      "epoch": 0.03,
       "learning_rate": 5e-07,
-      "loss": 1.1001,
+      "loss": 1.1046,
       "step": 10
     },
     {
-      "epoch": 0.02,
+      "epoch": 0.06,
       "learning_rate": 1e-06,
-      "loss": 1.0661,
+      "loss": 1.0664,
+      "step": 20
+    },
+    {
+      "epoch": 0.06,
+      "eval_accuracy": 0.44479243019924036,
+      "eval_combined_score": 0.3403229798934559,
+      "eval_f1": 0.2738667532127967,
+      "eval_loss": 1.065091609954834,
+      "eval_precision": 0.19784030596254612,
+      "eval_recall": 0.44479243019924036,
+      "eval_runtime": 49.6621,
+      "eval_samples_per_second": 604.364,
+      "eval_steps_per_second": 1.188,
       "step": 20
     },
     {
-      "epoch": 0.02,
+      "epoch": 0.09,
       "learning_rate": 1.5e-06,
-      "loss": 1.0782,
+      "loss": 1.0672,
       "step": 30
     },
     {
-      "epoch": 0.03,
+      "epoch": 0.12,
       "learning_rate": 2e-06,
-      "loss": 1.0578,
+      "loss": 1.0423,
       "step": 40
     },
     {
-      "epoch": 0.04,
-      "learning_rate": 2.5e-06,
-      "loss": 1.0395,
-      "step": 50
+      "epoch": 0.12,
+      "eval_accuracy": 0.5033650962883988,
+      "eval_combined_score": 0.4633375988688383,
+      "eval_f1": 0.4137746858205889,
+      "eval_loss": 1.0188277959823608,
+      "eval_precision": 0.43284551707796665,
+      "eval_recall": 0.5033650962883988,
+      "eval_runtime": 49.1985,
+      "eval_samples_per_second": 610.06,
+      "eval_steps_per_second": 1.199,
+      "step": 40
     },
     {
-      "epoch": 0.04,
-      "eval_accuracy": 0.4754781102152329,
-      "eval_loss": 1.0322505235671997,
-      "eval_runtime": 49.5848,
-      "eval_samples_per_second": 605.307,
-      "eval_steps_per_second": 1.19,
+      "epoch": 0.15,
+      "learning_rate": 2.5e-06,
+      "loss": 1.027,
       "step": 50
     },
     {
-      "epoch": 0.05,
+      "epoch": 0.18,
       "learning_rate": 3e-06,
-      "loss": 1.0308,
+      "loss": 1.0137,
+      "step": 60
+    },
+    {
+      "epoch": 0.18,
+      "eval_accuracy": 0.5279203038581995,
+      "eval_combined_score": 0.48433586826626607,
+      "eval_f1": 0.4596383010419703,
+      "eval_loss": 0.9871189594268799,
+      "eval_precision": 0.421864564306695,
+      "eval_recall": 0.5279203038581995,
+      "eval_runtime": 49.0237,
+      "eval_samples_per_second": 612.235,
+      "eval_steps_per_second": 1.204,
       "step": 60
     },
     {
-      "epoch": 0.05,
+      "epoch": 0.21,
       "learning_rate": 3.5e-06,
-      "loss": 1.0044,
+      "loss": 1.0005,
       "step": 70
     },
     {
-      "epoch": 0.06,
+      "epoch": 0.24,
       "learning_rate": 4e-06,
-      "loss": 1.0398,
+      "loss": 1.0027,
+      "step": 80
+    },
+    {
+      "epoch": 0.24,
+      "eval_accuracy": 0.5308189511561271,
+      "eval_combined_score": 0.4866197693698143,
+      "eval_f1": 0.46528335920671143,
+      "eval_loss": 0.9888613224029541,
+      "eval_precision": 0.4195578159602916,
+      "eval_recall": 0.5308189511561271,
+      "eval_runtime": 49.518,
+      "eval_samples_per_second": 606.124,
+      "eval_steps_per_second": 1.191,
       "step": 80
     },
     {
-      "epoch": 0.07,
+      "epoch": 0.27,
       "learning_rate": 4.5e-06,
-      "loss": 1.0116,
+      "loss": 0.9937,
       "step": 90
     },
     {
-      "epoch": 0.08,
+      "epoch": 0.3,
       "learning_rate": 5e-06,
-      "loss": 1.0067,
+      "loss": 0.9914,
       "step": 100
     },
     {
-      "epoch": 0.08,
-      "eval_accuracy": 0.5313187179316319,
-      "eval_loss": 0.9863238334655762,
-      "eval_runtime": 49.2356,
-      "eval_samples_per_second": 609.599,
-      "eval_steps_per_second": 1.198,
+      "epoch": 0.3,
+      "eval_accuracy": 0.5307856333710935,
+      "eval_combined_score": 0.48644275806230897,
+      "eval_f1": 0.4666072639999521,
+      "eval_loss": 0.9762536883354187,
+      "eval_precision": 0.4175925015070968,
+      "eval_recall": 0.5307856333710935,
+      "eval_runtime": 49.4629,
+      "eval_samples_per_second": 606.798,
+      "eval_steps_per_second": 1.193,
       "step": 100
     },
     {
-      "epoch": 0.08,
+      "epoch": 0.33,
       "learning_rate": 5.5e-06,
-      "loss": 1.0131,
+      "loss": 0.9884,
       "step": 110
     },
     {
-      "epoch": 0.09,
+      "epoch": 0.36,
       "learning_rate": 6e-06,
-      "loss": 1.0235,
+      "loss": 0.9826,
+      "step": 120
+    },
+    {
+      "epoch": 0.36,
+      "eval_accuracy": 0.5387819017791697,
+      "eval_combined_score": 0.4944622623129249,
+      "eval_f1": 0.47105988260831677,
+      "eval_loss": 0.9713281989097595,
+      "eval_precision": 0.4292253630850433,
+      "eval_recall": 0.5387819017791697,
+      "eval_runtime": 49.0682,
+      "eval_samples_per_second": 611.68,
+      "eval_steps_per_second": 1.202,
       "step": 120
     },
     {
-      "epoch": 0.1,
+      "epoch": 0.39,
       "learning_rate": 6.5e-06,
-      "loss": 1.0073,
+      "loss": 0.9882,
       "step": 130
     },
     {
-      "epoch": 0.11,
+      "epoch": 0.42,
       "learning_rate": 7e-06,
-      "loss": 0.9856,
+      "loss": 0.9788,
       "step": 140
     },
     {
-      "epoch": 0.11,
-      "learning_rate": 7.5e-06,
-      "loss": 1.0062,
-      "step": 150
+      "epoch": 0.42,
+      "eval_accuracy": 0.5312520823615646,
+      "eval_combined_score": 0.48712844254868687,
+      "eval_f1": 0.4673598197970026,
+      "eval_loss": 0.9766249656677246,
+      "eval_precision": 0.41864978567461564,
+      "eval_recall": 0.5312520823615646,
+      "eval_runtime": 49.1859,
+      "eval_samples_per_second": 610.215,
+      "eval_steps_per_second": 1.2,
+      "step": 140
     },
     {
-      "epoch": 0.11,
-      "eval_accuracy": 0.5358499366962084,
-      "eval_loss": 0.9735665917396545,
-      "eval_runtime": 49.4568,
-      "eval_samples_per_second": 606.874,
-      "eval_steps_per_second": 1.193,
+      "epoch": 0.45,
+      "learning_rate": 7.5e-06,
+      "loss": 0.9854,
       "step": 150
     },
     {
-      "epoch": 0.12,
+      "epoch": 0.48,
       "learning_rate": 8e-06,
-      "loss": 0.9622,
+      "loss": 0.984,
+      "step": 160
+    },
+    {
+      "epoch": 0.48,
+      "eval_accuracy": 0.5398147531152129,
+      "eval_combined_score": 0.49477075635813,
+      "eval_f1": 0.4751263005883661,
+      "eval_loss": 0.9589501619338989,
+      "eval_precision": 0.4243272186137281,
+      "eval_recall": 0.5398147531152129,
+      "eval_runtime": 49.0847,
+      "eval_samples_per_second": 611.473,
+      "eval_steps_per_second": 1.202,
       "step": 160
     },
     {
-      "epoch": 0.13,
+      "epoch": 0.51,
       "learning_rate": 8.500000000000002e-06,
-      "loss": 0.9583,
+      "loss": 0.9795,
       "step": 170
     },
     {
-      "epoch": 0.14,
+      "epoch": 0.54,
       "learning_rate": 9e-06,
-      "loss": 0.987,
+      "loss": 0.9694,
+      "step": 180
+    },
+    {
+      "epoch": 0.54,
+      "eval_accuracy": 0.5423469047777704,
+      "eval_combined_score": 0.49718028497336725,
+      "eval_f1": 0.4771728160733735,
+      "eval_loss": 0.953514814376831,
+      "eval_precision": 0.42685451426455484,
+      "eval_recall": 0.5423469047777704,
+      "eval_runtime": 49.0032,
+      "eval_samples_per_second": 612.491,
+      "eval_steps_per_second": 1.204,
       "step": 180
     },
     {
-      "epoch": 0.14,
+      "epoch": 0.57,
       "learning_rate": 9.5e-06,
-      "loss": 0.9995,
+      "loss": 0.9757,
       "step": 190
     },
     {
-      "epoch": 0.15,
+      "epoch": 0.6,
       "learning_rate": 1e-05,
-      "loss": 0.9953,
+      "loss": 0.9676,
       "step": 200
     },
     {
-      "epoch": 0.15,
-      "eval_accuracy": 0.5406476977410541,
-      "eval_loss": 0.9847397804260254,
-      "eval_runtime": 49.5469,
-      "eval_samples_per_second": 605.769,
-      "eval_steps_per_second": 1.191,
+      "epoch": 0.6,
+      "eval_accuracy": 0.567201972412874,
+      "eval_combined_score": 0.5200447629299639,
+      "eval_f1": 0.49911443945569844,
+      "eval_loss": 0.9273685812950134,
+      "eval_precision": 0.44666066743840943,
+      "eval_recall": 0.567201972412874,
+      "eval_runtime": 49.1912,
+      "eval_samples_per_second": 610.15,
+      "eval_steps_per_second": 1.199,
       "step": 200
     },
     {
-      "epoch": 0.16,
+      "epoch": 0.63,
       "learning_rate": 1.0500000000000001e-05,
-      "loss": 0.9894,
+      "loss": 0.9708,
       "step": 210
     },
     {
-      "epoch": 0.17,
+      "epoch": 0.66,
       "learning_rate": 1.1e-05,
-      "loss": 0.9857,
+      "loss": 0.9753,
       "step": 220
     },
     {
-      "epoch": 0.17,
+      "epoch": 0.66,
+      "eval_accuracy": 0.573598987139335,
+      "eval_combined_score": 0.5278568904198743,
+      "eval_f1": 0.5026102591641352,
+      "eval_loss": 0.912590503692627,
+      "eval_precision": 0.4616193282366919,
+      "eval_recall": 0.573598987139335,
+      "eval_runtime": 49.2154,
+      "eval_samples_per_second": 609.849,
+      "eval_steps_per_second": 1.199,
+      "step": 220
+    },
+    {
+      "epoch": 0.69,
       "learning_rate": 1.15e-05,
-      "loss": 0.9806,
+      "loss": 0.9586,
       "step": 230
     },
     {
-      "epoch": 0.18,
+      "epoch": 0.72,
       "learning_rate": 1.2e-05,
-      "loss": 0.9802,
+      "loss": 0.9557,
       "step": 240
     },
     {
-      "epoch": 0.19,
-      "learning_rate": 1.25e-05,
-      "loss": 0.9559,
-      "step": 250
+      "epoch": 0.72,
+      "eval_accuracy": 0.5759645498767242,
+      "eval_combined_score": 0.5280177627115059,
+      "eval_f1": 0.5069205876947673,
+      "eval_loss": 0.9052607417106628,
+      "eval_precision": 0.45322136339780783,
+      "eval_recall": 0.5759645498767242,
+      "eval_runtime": 49.1399,
+      "eval_samples_per_second": 610.787,
+      "eval_steps_per_second": 1.201,
+      "step": 240
     },
     {
-      "epoch": 0.19,
-      "eval_accuracy": 0.5136269740787632,
-      "eval_loss": 1.0206754207611084,
-      "eval_runtime": 49.5108,
-      "eval_samples_per_second": 606.212,
-      "eval_steps_per_second": 1.192,
+      "epoch": 0.75,
+      "learning_rate": 1.25e-05,
+      "loss": 0.9518,
       "step": 250
     },
     {
-      "epoch": 0.2,
+      "epoch": 0.78,
       "learning_rate": 1.3e-05,
-      "loss": 0.9877,
+      "loss": 0.9508,
       "step": 260
     },
     {
-      "epoch": 0.2,
+      "epoch": 0.78,
+      "eval_accuracy": 0.5766975411474645,
+      "eval_combined_score": 0.534067292009198,
+      "eval_f1": 0.5018221478654629,
+      "eval_loss": 0.9178985953330994,
+      "eval_precision": 0.4810519378764,
+      "eval_recall": 0.5766975411474645,
+      "eval_runtime": 49.2092,
+      "eval_samples_per_second": 609.927,
+      "eval_steps_per_second": 1.199,
+      "step": 260
+    },
+    {
+      "epoch": 0.81,
       "learning_rate": 1.35e-05,
-      "loss": 0.9878,
+      "loss": 0.9405,
       "step": 270
     },
     {
-      "epoch": 0.21,
+      "epoch": 0.84,
       "learning_rate": 1.4e-05,
-      "loss": 0.9646,
+      "loss": 0.9355,
+      "step": 280
+    },
+    {
+      "epoch": 0.84,
+      "eval_accuracy": 0.58915839275005,
+      "eval_combined_score": 0.5407092882328752,
+      "eval_f1": 0.5183161389995846,
+      "eval_loss": 0.8937407732009888,
+      "eval_precision": 0.46620422843181647,
+      "eval_recall": 0.58915839275005,
+      "eval_runtime": 49.2523,
+      "eval_samples_per_second": 609.393,
+      "eval_steps_per_second": 1.198,
       "step": 280
     },
     {
-      "epoch": 0.22,
+      "epoch": 0.87,
       "learning_rate": 1.4500000000000002e-05,
-      "loss": 0.9598,
+      "loss": 0.9235,
       "step": 290
     },
     {
-      "epoch": 0.23,
+      "epoch": 0.9,
       "learning_rate": 1.5e-05,
-      "loss": 0.9567,
+      "loss": 0.9,
       "step": 300
     },
     {
-      "epoch": 0.23,
-      "eval_accuracy": 0.5693676284400613,
-      "eval_loss": 0.9298574924468994,
-      "eval_runtime": 49.5798,
-      "eval_samples_per_second": 605.367,
-      "eval_steps_per_second": 1.19,
+      "epoch": 0.9,
+      "eval_accuracy": 0.613013926834144,
+      "eval_combined_score": 0.5923759012239074,
+      "eval_f1": 0.5579282363395321,
+      "eval_loss": 0.8468813300132751,
+      "eval_precision": 0.5855475148878095,
+      "eval_recall": 0.613013926834144,
+      "eval_runtime": 49.1896,
+      "eval_samples_per_second": 610.17,
+      "eval_steps_per_second": 1.199,
       "step": 300
     },
     {
-      "epoch": 0.23,
+      "epoch": 0.93,
       "learning_rate": 1.55e-05,
-      "loss": 0.9659,
+      "loss": 0.8857,
       "step": 310
     },
     {
-      "epoch": 0.24,
+      "epoch": 0.96,
       "learning_rate": 1.6e-05,
-      "loss": 0.9803,
+      "loss": 0.993,
+      "step": 320
+    },
+    {
+      "epoch": 0.96,
+      "eval_accuracy": 0.6046511627906976,
+      "eval_combined_score": 0.590510828847508,
+      "eval_f1": 0.5352474275558012,
+      "eval_loss": 0.8615403771400452,
+      "eval_precision": 0.6174935622528357,
+      "eval_recall": 0.6046511627906976,
+      "eval_runtime": 49.37,
+      "eval_samples_per_second": 607.94,
+      "eval_steps_per_second": 1.195,
       "step": 320
     },
     {
-      "epoch": 0.25,
+      "epoch": 0.99,
       "learning_rate": 1.65e-05,
-      "loss": 0.959,
+      "loss": 0.8418,
       "step": 330
     },
     {
-      "epoch": 0.26,
+      "epoch": 1.02,
       "learning_rate": 1.7000000000000003e-05,
-      "loss": 0.9298,
+      "loss": 0.8527,
       "step": 340
     },
     {
-      "epoch": 0.26,
-      "learning_rate": 1.7500000000000002e-05,
-      "loss": 0.8923,
-      "step": 350
+      "epoch": 1.02,
+      "eval_accuracy": 0.6439328313453722,
+      "eval_combined_score": 0.6332615383344228,
+      "eval_f1": 0.6212200513859825,
+      "eval_loss": 0.7896137237548828,
+      "eval_precision": 0.6239604392609642,
+      "eval_recall": 0.6439328313453722,
+      "eval_runtime": 49.1146,
+      "eval_samples_per_second": 611.102,
+      "eval_steps_per_second": 1.201,
+      "step": 340
     },
     {
-      "epoch": 0.26,
-      "eval_accuracy": 0.603285133604318,
-      "eval_loss": 0.8731092810630798,
-      "eval_runtime": 49.5126,
-      "eval_samples_per_second": 606.189,
-      "eval_steps_per_second": 1.192,
+      "epoch": 1.05,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 0.8982,
       "step": 350
     },
     {
-      "epoch": 0.27,
+      "epoch": 1.08,
       "learning_rate": 1.8e-05,
-      "loss": 0.9255,
+      "loss": 0.966,
       "step": 360
     },
     {
-      "epoch": 0.28,
+      "epoch": 1.08,
+      "eval_accuracy": 0.5315852602119011,
+      "eval_combined_score": 0.5500800253024862,
+      "eval_f1": 0.4943799031015135,
+      "eval_loss": 1.0123510360717773,
+      "eval_precision": 0.642769677684629,
+      "eval_recall": 0.5315852602119011,
+      "eval_runtime": 49.1941,
+      "eval_samples_per_second": 610.114,
+      "eval_steps_per_second": 1.199,
+      "step": 360
+    },
+    {
+      "epoch": 1.11,
       "learning_rate": 1.85e-05,
-      "loss": 0.9567,
+      "loss": 0.9413,
       "step": 370
     },
     {
-      "epoch": 0.29,
+      "epoch": 1.14,
       "learning_rate": 1.9e-05,
-      "loss": 0.9135,
+      "loss": 0.8441,
+      "step": 380
+    },
+    {
+      "epoch": 1.14,
+      "eval_accuracy": 0.6488638635303525,
+      "eval_combined_score": 0.648329789222037,
+      "eval_f1": 0.6371420074616733,
+      "eval_loss": 0.791083574295044,
+      "eval_precision": 0.6584494223657698,
+      "eval_recall": 0.6488638635303525,
+      "eval_runtime": 49.3484,
+      "eval_samples_per_second": 608.207,
+      "eval_steps_per_second": 1.196,
       "step": 380
     },
     {
-      "epoch": 0.29,
+      "epoch": 1.17,
       "learning_rate": 1.95e-05,
-      "loss": 0.8978,
+      "loss": 0.8223,
       "step": 390
     },
     {
-      "epoch": 0.3,
+      "epoch": 1.2,
       "learning_rate": 2e-05,
-      "loss": 0.9617,
+      "loss": 0.8226,
       "step": 400
     },
     {
-      "epoch": 0.3,
-      "eval_accuracy": 0.5401479309655494,
-      "eval_loss": 0.9500759840011597,
-      "eval_runtime": 49.5248,
-      "eval_samples_per_second": 606.04,
-      "eval_steps_per_second": 1.191,
+      "epoch": 1.2,
+      "eval_accuracy": 0.6699873392416872,
+      "eval_combined_score": 0.6621103162794577,
+      "eval_f1": 0.6459453754294268,
+      "eval_loss": 0.7472424507141113,
+      "eval_precision": 0.6625212112050299,
+      "eval_recall": 0.6699873392416872,
+      "eval_runtime": 49.3153,
+      "eval_samples_per_second": 608.614,
+      "eval_steps_per_second": 1.196,
       "step": 400
     },
     {
-      "epoch": 0.31,
+      "epoch": 1.23,
       "learning_rate": 2.05e-05,
-      "loss": 0.9387,
+      "loss": 0.7924,
       "step": 410
     },
     {
-      "epoch": 0.32,
+      "epoch": 1.26,
       "learning_rate": 2.1000000000000002e-05,
-      "loss": 0.8832,
+      "loss": 0.7948,
+      "step": 420
+    },
+    {
+      "epoch": 1.26,
+      "eval_accuracy": 0.6580595721996402,
+      "eval_combined_score": 0.6513132521613654,
+      "eval_f1": 0.6095365334747936,
+      "eval_loss": 0.7663838863372803,
+      "eval_precision": 0.6795973307713873,
+      "eval_recall": 0.6580595721996402,
+      "eval_runtime": 49.2143,
+      "eval_samples_per_second": 609.863,
+      "eval_steps_per_second": 1.199,
       "step": 420
     },
     {
-      "epoch": 0.32,
+      "epoch": 1.29,
       "learning_rate": 2.1499999999999997e-05,
-      "loss": 0.9087,
+      "loss": 0.776,
       "step": 430
     },
     {
-      "epoch": 0.33,
+      "epoch": 1.32,
       "learning_rate": 2.2e-05,
-      "loss": 0.9192,
+      "loss": 0.7428,
       "step": 440
     },
     {
-      "epoch": 0.34,
-      "learning_rate": 2.2499999999999998e-05,
-      "loss": 0.8456,
-      "step": 450
+      "epoch": 1.32,
+      "eval_accuracy": 0.6991737189311654,
+      "eval_combined_score": 0.6905198116505921,
+      "eval_f1": 0.653333567856404,
+      "eval_loss": 0.6993714570999146,
+      "eval_precision": 0.7103982408836337,
+      "eval_recall": 0.6991737189311654,
+      "eval_runtime": 49.1465,
+      "eval_samples_per_second": 610.704,
+      "eval_steps_per_second": 1.2,
+      "step": 440
     },
     {
-      "epoch": 0.34,
-      "eval_accuracy": 0.6245085626707536,
-      "eval_loss": 0.8298193216323853,
-      "eval_runtime": 49.5226,
-      "eval_samples_per_second": 606.066,
-      "eval_steps_per_second": 1.191,
+      "epoch": 1.35,
+      "learning_rate": 2.2499999999999998e-05,
+      "loss": 0.7512,
       "step": 450
     },
     {
-      "epoch": 0.35,
+      "epoch": 1.38,
       "learning_rate": 2.3e-05,
-      "loss": 0.8511,
+      "loss": 0.7109,
       "step": 460
     },
     {
-      "epoch": 0.35,
+      "epoch": 1.38,
+      "eval_accuracy": 0.7283600986206437,
+      "eval_combined_score": 0.7200537958840363,
+      "eval_f1": 0.6890059409106147,
+      "eval_loss": 0.6510820984840393,
+      "eval_precision": 0.7344890453842432,
+      "eval_recall": 0.7283600986206437,
+      "eval_runtime": 49.0446,
+      "eval_samples_per_second": 611.974,
+      "eval_steps_per_second": 1.203,
+      "step": 460
+    },
+    {
+      "epoch": 1.41,
       "learning_rate": 2.3500000000000002e-05,
-      "loss": 0.9211,
+      "loss": 0.701,
       "step": 470
     },
     {
-      "epoch": 0.36,
+      "epoch": 1.44,
       "learning_rate": 2.4e-05,
-      "loss": 0.8416,
+      "loss": 0.6882,
       "step": 480
     },
     {
-      "epoch": 0.37,
+      "epoch": 1.44,
+      "eval_accuracy": 0.7576797494502565,
+      "eval_combined_score": 0.7514829004826655,
+      "eval_f1": 0.7336051419487455,
+      "eval_loss": 0.5987845063209534,
+      "eval_precision": 0.7569669610814035,
+      "eval_recall": 0.7576797494502565,
+      "eval_runtime": 49.3939,
+      "eval_samples_per_second": 607.646,
+      "eval_steps_per_second": 1.194,
+      "step": 480
+    },
+    {
+      "epoch": 1.47,
       "learning_rate": 2.4500000000000003e-05,
-      "loss": 0.849,
+      "loss": 0.651,
       "step": 490
     },
     {
-      "epoch": 0.38,
+      "epoch": 1.5,
       "learning_rate": 2.5e-05,
-      "loss": 0.8098,
+      "loss": 0.7296,
       "step": 500
     },
     {
-      "epoch": 0.38,
-      "eval_accuracy": 0.6523622309588859,
-      "eval_loss": 0.771039605140686,
-      "eval_runtime": 49.4859,
-      "eval_samples_per_second": 606.517,
-      "eval_steps_per_second": 1.192,
+      "epoch": 1.5,
+      "eval_accuracy": 0.7564136736189778,
+      "eval_combined_score": 0.7674838523650971,
+      "eval_f1": 0.766902713706533,
+      "eval_loss": 0.5992804765701294,
+      "eval_precision": 0.7902053485158999,
+      "eval_recall": 0.7564136736189778,
+      "eval_runtime": 49.1622,
+      "eval_samples_per_second": 610.509,
+      "eval_steps_per_second": 1.2,
       "step": 500
     },
     {
-      "epoch": 0.38,
+      "epoch": 1.53,
       "learning_rate": 2.55e-05,
-      "loss": 0.8406,
+      "loss": 0.5943,
       "step": 510
     },
     {
-      "epoch": 0.39,
+      "epoch": 1.57,
       "learning_rate": 2.6e-05,
-      "loss": 0.8868,
+      "loss": 0.5677,
+      "step": 520
+    },
+    {
+      "epoch": 1.57,
+      "eval_accuracy": 0.8125874591857133,
+      "eval_combined_score": 0.8104607682202668,
+      "eval_f1": 0.7941544044601274,
+      "eval_loss": 0.5067983865737915,
+      "eval_precision": 0.8225137500495135,
+      "eval_recall": 0.8125874591857133,
+      "eval_runtime": 49.0571,
+      "eval_samples_per_second": 611.817,
+      "eval_steps_per_second": 1.203,
       "step": 520
     },
     {
-      "epoch": 0.4,
+      "epoch": 1.6,
       "learning_rate": 2.65e-05,
-      "loss": 0.8206,
+      "loss": 0.5818,
       "step": 530
     },
     {
-      "epoch": 0.41,
+      "epoch": 1.63,
       "learning_rate": 2.7e-05,
-      "loss": 0.7607,
+      "loss": 0.5096,
       "step": 540
     },
     {
-      "epoch": 0.41,
-      "learning_rate": 2.75e-05,
-      "loss": 0.6892,
-      "step": 550
+      "epoch": 1.63,
+      "eval_accuracy": 0.8520023988805224,
+      "eval_combined_score": 0.8500052190439381,
+      "eval_f1": 0.8448837390352126,
+      "eval_loss": 0.42726200819015503,
+      "eval_precision": 0.8511323393794948,
+      "eval_recall": 0.8520023988805224,
+      "eval_runtime": 49.0407,
+      "eval_samples_per_second": 612.023,
+      "eval_steps_per_second": 1.203,
+      "step": 540
     },
     {
-      "epoch": 0.41,
-      "eval_accuracy": 0.7312254281335376,
-      "eval_loss": 0.670229971408844,
-      "eval_runtime": 49.5662,
-      "eval_samples_per_second": 605.533,
-      "eval_steps_per_second": 1.19,
+      "epoch": 1.66,
+      "learning_rate": 2.75e-05,
+      "loss": 0.462,
       "step": 550
     },
     {
-      "epoch": 0.42,
+      "epoch": 1.69,
       "learning_rate": 2.8e-05,
-      "loss": 0.7065,
+      "loss": 0.4452,
+      "step": 560
+    },
+    {
+      "epoch": 1.69,
+      "eval_accuracy": 0.8721929766109149,
+      "eval_combined_score": 0.8711433017214012,
+      "eval_f1": 0.8698654313719106,
+      "eval_loss": 0.37960025668144226,
+      "eval_precision": 0.8703218222918644,
+      "eval_recall": 0.8721929766109149,
+      "eval_runtime": 49.6519,
+      "eval_samples_per_second": 604.489,
+      "eval_steps_per_second": 1.188,
       "step": 560
     },
     {
-      "epoch": 0.43,
+      "epoch": 1.72,
       "learning_rate": 2.85e-05,
-      "loss": 0.7075,
+      "loss": 0.4267,
       "step": 570
     },
     {
-      "epoch": 0.44,
+      "epoch": 1.75,
       "learning_rate": 2.9000000000000004e-05,
-      "loss": 0.6793,
+      "loss": 0.3836,
       "step": 580
     },
     {
-      "epoch": 0.44,
+      "epoch": 1.75,
+      "eval_accuracy": 0.8756580262544146,
+      "eval_combined_score": 0.876232023467392,
+      "eval_f1": 0.8758457824110393,
+      "eval_loss": 0.3854842483997345,
+      "eval_precision": 0.8777662589496996,
+      "eval_recall": 0.8756580262544146,
+      "eval_runtime": 49.2229,
+      "eval_samples_per_second": 609.757,
+      "eval_steps_per_second": 1.199,
+      "step": 580
+    },
+    {
+      "epoch": 1.78,
       "learning_rate": 2.95e-05,
-      "loss": 0.6057,
+      "loss": 0.3828,
       "step": 590
     },
     {
-      "epoch": 0.45,
+      "epoch": 1.81,
       "learning_rate": 3e-05,
-      "loss": 0.5794,
+      "loss": 0.3783,
       "step": 600
     },
     {
-      "epoch": 0.45,
-      "eval_accuracy": 0.7694409275671353,
-      "eval_loss": 0.6066040396690369,
-      "eval_runtime": 49.5056,
-      "eval_samples_per_second": 606.275,
-      "eval_steps_per_second": 1.192,
+      "epoch": 1.81,
+      "eval_accuracy": 0.8893849536882789,
+      "eval_combined_score": 0.8889004932680924,
+      "eval_f1": 0.8883136621852403,
+      "eval_loss": 0.35855188965797424,
+      "eval_precision": 0.8885184035105714,
+      "eval_recall": 0.8893849536882789,
+      "eval_runtime": 49.2611,
+      "eval_samples_per_second": 609.285,
+      "eval_steps_per_second": 1.198,
       "step": 600
     },
     {
-      "epoch": 0.46,
+      "epoch": 1.84,
       "learning_rate": 3.05e-05,
-      "loss": 0.609,
+      "loss": 0.3466,
       "step": 610
     },
     {
-      "epoch": 0.47,
+      "epoch": 1.87,
       "learning_rate": 3.1e-05,
-      "loss": 0.6058,
+      "loss": 0.3496,
       "step": 620
     },
     {
-      "epoch": 0.47,
+      "epoch": 1.87,
+      "eval_accuracy": 0.8971813153861531,
+      "eval_combined_score": 0.8972621407363919,
+      "eval_f1": 0.8972738562660447,
+      "eval_loss": 0.321043998003006,
+      "eval_precision": 0.8974120759072168,
+      "eval_recall": 0.8971813153861531,
+      "eval_runtime": 49.4334,
+      "eval_samples_per_second": 607.16,
+      "eval_steps_per_second": 1.194,
+      "step": 620
+    },
+    {
+      "epoch": 1.9,
       "learning_rate": 3.15e-05,
-      "loss": 0.5518,
+      "loss": 0.3421,
       "step": 630
     },
     {
-      "epoch": 0.48,
+      "epoch": 1.93,
       "learning_rate": 3.2e-05,
-      "loss": 0.6132,
+      "loss": 0.3585,
       "step": 640
     },
     {
-      "epoch": 0.49,
-      "learning_rate": 3.2500000000000004e-05,
-      "loss": 0.5262,
-      "step": 650
+      "epoch": 1.93,
+      "eval_accuracy": 0.9035450123275804,
+      "eval_combined_score": 0.9033726098964578,
+      "eval_f1": 0.9030761902479666,
+      "eval_loss": 0.3006099760532379,
+      "eval_precision": 0.903324224682704,
+      "eval_recall": 0.9035450123275804,
+      "eval_runtime": 49.1842,
+      "eval_samples_per_second": 610.237,
+      "eval_steps_per_second": 1.2,
+      "step": 640
     },
     {
-      "epoch": 0.49,
-      "eval_accuracy": 0.8218498034250683,
-      "eval_loss": 0.4961170256137848,
-      "eval_runtime": 49.5394,
-      "eval_samples_per_second": 605.861,
-      "eval_steps_per_second": 1.191,
+      "epoch": 1.96,
+      "learning_rate": 3.2500000000000004e-05,
+      "loss": 0.3442,
       "step": 650
     },
     {
-      "epoch": 0.5,
+      "epoch": 1.99,
       "learning_rate": 3.3e-05,
-      "loss": 0.5355,
+      "loss": 0.345,
+      "step": 660
+    },
+    {
+      "epoch": 1.99,
+      "eval_accuracy": 0.9014126740854268,
+      "eval_combined_score": 0.9026415340332592,
+      "eval_f1": 0.9025066760774267,
+      "eval_loss": 0.30540063977241516,
+      "eval_precision": 0.9052341118847564,
+      "eval_recall": 0.9014126740854268,
+      "eval_runtime": 49.4063,
+      "eval_samples_per_second": 607.494,
+      "eval_steps_per_second": 1.194,
       "step": 660
     },
     {
-      "epoch": 0.5,
+      "epoch": 2.02,
       "learning_rate": 3.35e-05,
-      "loss": 0.4905,
+      "loss": 0.3246,
       "step": 670
     },
     {
-      "epoch": 0.51,
+      "epoch": 2.05,
       "learning_rate": 3.4000000000000007e-05,
-      "loss": 0.4992,
+      "loss": 0.3327,
+      "step": 680
+    },
+    {
+      "epoch": 2.05,
+      "eval_accuracy": 0.8912507496501633,
+      "eval_combined_score": 0.8911549520027604,
+      "eval_f1": 0.8866402874920362,
+      "eval_loss": 0.3174073100090027,
+      "eval_precision": 0.895478021218679,
+      "eval_recall": 0.8912507496501633,
+      "eval_runtime": 49.3463,
+      "eval_samples_per_second": 608.232,
+      "eval_steps_per_second": 1.196,
       "step": 680
     },
     {
-      "epoch": 0.52,
+      "epoch": 2.08,
       "learning_rate": 3.4500000000000005e-05,
-      "loss": 0.4959,
+      "loss": 0.3172,
       "step": 690
     },
     {
-      "epoch": 0.53,
+      "epoch": 2.11,
       "learning_rate": 3.5000000000000004e-05,
-      "loss": 0.4693,
+      "loss": 0.2962,
       "step": 700
     },
     {
-      "epoch": 0.53,
-      "eval_accuracy": 0.851469314319984,
-      "eval_loss": 0.43606042861938477,
-      "eval_runtime": 49.5557,
-      "eval_samples_per_second": 605.662,
-      "eval_steps_per_second": 1.191,
+      "epoch": 2.11,
+      "eval_accuracy": 0.9122076364363297,
+      "eval_combined_score": 0.9124808443428073,
+      "eval_f1": 0.9125293911738805,
+      "eval_loss": 0.2769884169101715,
+      "eval_precision": 0.9129787133246888,
+      "eval_recall": 0.9122076364363297,
+      "eval_runtime": 49.3634,
+      "eval_samples_per_second": 608.021,
+      "eval_steps_per_second": 1.195,
       "step": 700
     },
     {
-      "epoch": 0.53,
+      "epoch": 2.14,
       "learning_rate": 3.5499999999999996e-05,
-      "loss": 0.4725,
+      "loss": 0.287,
       "step": 710
     },
     {
-      "epoch": 0.54,
+      "epoch": 2.17,
       "learning_rate": 3.6e-05,
-      "loss": 0.5033,
+      "loss": 0.3032,
       "step": 720
     },
     {
-      "epoch": 0.55,
+      "epoch": 2.17,
+      "eval_accuracy": 0.9062437529153062,
+      "eval_combined_score": 0.9068222505646787,
+      "eval_f1": 0.9054896753410417,
+      "eval_loss": 0.2979203760623932,
+      "eval_precision": 0.9093118210870608,
+      "eval_recall": 0.9062437529153062,
+      "eval_runtime": 49.0419,
+      "eval_samples_per_second": 612.007,
+      "eval_steps_per_second": 1.203,
+      "step": 720
+    },
+    {
+      "epoch": 2.2,
       "learning_rate": 3.65e-05,
-      "loss": 0.4679,
+      "loss": 0.3017,
       "step": 730
     },
     {
-      "epoch": 0.56,
+      "epoch": 2.23,
       "learning_rate": 3.7e-05,
-      "loss": 0.4639,
+      "loss": 0.27,
       "step": 740
     },
     {
-      "epoch": 0.56,
-      "learning_rate": 3.75e-05,
-      "loss": 0.434,
-      "step": 750
+      "epoch": 2.23,
+      "eval_accuracy": 0.8997801026187779,
+      "eval_combined_score": 0.9003031124352014,
+      "eval_f1": 0.8971406374938125,
+      "eval_loss": 0.29734131693840027,
+      "eval_precision": 0.9045116070094374,
+      "eval_recall": 0.8997801026187779,
+      "eval_runtime": 49.1818,
+      "eval_samples_per_second": 610.267,
+      "eval_steps_per_second": 1.2,
+      "step": 740
     },
     {
-      "epoch": 0.56,
-      "eval_accuracy": 0.8657293263143866,
-      "eval_loss": 0.41026708483695984,
-      "eval_runtime": 49.5112,
-      "eval_samples_per_second": 606.206,
-      "eval_steps_per_second": 1.192,
+      "epoch": 2.26,
+      "learning_rate": 3.75e-05,
+      "loss": 0.3042,
       "step": 750
     },
     {
-      "epoch": 0.57,
+      "epoch": 2.29,
       "learning_rate": 3.8e-05,
-      "loss": 0.3815,
+      "loss": 0.2912,
+      "step": 760
+    },
+    {
+      "epoch": 2.29,
+      "eval_accuracy": 0.9221696541613914,
+      "eval_combined_score": 0.9221674217032323,
+      "eval_f1": 0.922071694968714,
+      "eval_loss": 0.24671417474746704,
+      "eval_precision": 0.9222586835214323,
+      "eval_recall": 0.9221696541613914,
+      "eval_runtime": 49.2126,
+      "eval_samples_per_second": 609.884,
+      "eval_steps_per_second": 1.199,
       "step": 760
     },
     {
-      "epoch": 0.58,
+      "epoch": 2.32,
       "learning_rate": 3.85e-05,
-      "loss": 0.4073,
+      "loss": 0.2609,
       "step": 770
     },
     {
-      "epoch": 0.59,
+      "epoch": 2.35,
       "learning_rate": 3.9e-05,
-      "loss": 0.424,
+      "loss": 0.2412,
+      "step": 780
+    },
+    {
+      "epoch": 2.35,
+      "eval_accuracy": 0.9113080562404211,
+      "eval_combined_score": 0.9131963179862455,
+      "eval_f1": 0.9128361841568111,
+      "eval_loss": 0.2760636806488037,
+      "eval_precision": 0.917332975307328,
+      "eval_recall": 0.9113080562404211,
+      "eval_runtime": 49.2008,
+      "eval_samples_per_second": 610.031,
+      "eval_steps_per_second": 1.199,
       "step": 780
     },
     {
-      "epoch": 0.59,
+      "epoch": 2.38,
       "learning_rate": 3.95e-05,
-      "loss": 0.3867,
+      "loss": 0.2755,
       "step": 790
     },
     {
-      "epoch": 0.6,
+      "epoch": 2.41,
       "learning_rate": 4e-05,
-      "loss": 0.463,
+      "loss": 0.2746,
       "step": 800
     },
     {
-      "epoch": 0.6,
-      "eval_accuracy": 0.8671286732858,
-      "eval_loss": 0.3895261585712433,
-      "eval_runtime": 49.5727,
-      "eval_samples_per_second": 605.455,
-      "eval_steps_per_second": 1.19,
+      "epoch": 2.41,
+      "eval_accuracy": 0.9260011994402612,
+      "eval_combined_score": 0.9259453001449679,
+      "eval_f1": 0.9257439757889774,
+      "eval_loss": 0.24102580547332764,
+      "eval_precision": 0.9260348259103722,
+      "eval_recall": 0.9260011994402612,
+      "eval_runtime": 49.2957,
+      "eval_samples_per_second": 608.856,
+      "eval_steps_per_second": 1.197,
       "step": 800
     },
     {
-      "epoch": 0.61,
+      "epoch": 2.44,
       "learning_rate": 4.05e-05,
-      "loss": 0.407,
+      "loss": 0.2563,
       "step": 810
     },
     {
-      "epoch": 0.62,
+      "epoch": 2.47,
       "learning_rate": 4.1e-05,
-      "loss": 0.4077,
+      "loss": 0.2637,
       "step": 820
     },
     {
-      "epoch": 0.62,
+      "epoch": 2.47,
+      "eval_accuracy": 0.922103018591324,
+      "eval_combined_score": 0.9220769784769673,
+      "eval_f1": 0.9212994658547122,
+      "eval_loss": 0.2446586638689041,
+      "eval_precision": 0.9228024108705088,
+      "eval_recall": 0.922103018591324,
+      "eval_runtime": 48.9946,
+      "eval_samples_per_second": 612.598,
+      "eval_steps_per_second": 1.204,
+      "step": 820
+    },
+    {
+      "epoch": 2.5,
       "learning_rate": 4.1500000000000006e-05,
-      "loss": 0.4015,
+      "loss": 0.2492,
       "step": 830
     },
     {
-      "epoch": 0.63,
+      "epoch": 2.53,
       "learning_rate": 4.2000000000000004e-05,
-      "loss": 0.3652,
+      "loss": 0.2605,
       "step": 840
     },
     {
-      "epoch": 0.64,
-      "learning_rate": 4.25e-05,
-      "loss": 0.411,
-      "step": 850
+      "epoch": 2.53,
+      "eval_accuracy": 0.9237355900579729,
+      "eval_combined_score": 0.9240135925879284,
+      "eval_f1": 0.9231767824872119,
+      "eval_loss": 0.2475174069404602,
+      "eval_precision": 0.9254064077485559,
+      "eval_recall": 0.9237355900579729,
+      "eval_runtime": 49.3973,
+      "eval_samples_per_second": 607.604,
+      "eval_steps_per_second": 1.194,
+      "step": 840
     },
     {
-      "epoch": 0.64,
-      "eval_accuracy": 0.8794895715332844,
-      "eval_loss": 0.36858388781547546,
-      "eval_runtime": 49.5593,
-      "eval_samples_per_second": 605.618,
-      "eval_steps_per_second": 1.19,
+      "epoch": 2.56,
+      "learning_rate": 4.25e-05,
+      "loss": 0.2393,
       "step": 850
     },
     {
-      "epoch": 0.65,
+      "epoch": 2.59,
       "learning_rate": 4.2999999999999995e-05,
-      "loss": 0.4162,
+      "loss": 0.2517,
+      "step": 860
+    },
+    {
+      "epoch": 2.59,
+      "eval_accuracy": 0.9265342840007996,
+      "eval_combined_score": 0.9265368337935618,
+      "eval_f1": 0.9258759190762016,
+      "eval_loss": 0.2590079605579376,
+      "eval_precision": 0.927202848096446,
+      "eval_recall": 0.9265342840007996,
+      "eval_runtime": 48.961,
+      "eval_samples_per_second": 613.019,
+      "eval_steps_per_second": 1.205,
       "step": 860
     },
     {
-      "epoch": 0.65,
+      "epoch": 2.62,
       "learning_rate": 4.35e-05,
-      "loss": 0.3985,
+      "loss": 0.2515,
       "step": 870
     },
     {
-      "epoch": 0.66,
+      "epoch": 2.65,
       "learning_rate": 4.4e-05,
-      "loss": 0.3925,
+      "loss": 0.2453,
+      "step": 880
+    },
+    {
+      "epoch": 2.65,
+      "eval_accuracy": 0.930032651429333,
+      "eval_combined_score": 0.9305237768064085,
+      "eval_f1": 0.9305263249699511,
+      "eval_loss": 0.22479340434074402,
+      "eval_precision": 0.9315034793970168,
+      "eval_recall": 0.930032651429333,
+      "eval_runtime": 49.2223,
+      "eval_samples_per_second": 609.764,
+      "eval_steps_per_second": 1.199,
       "step": 880
     },
     {
-      "epoch": 0.67,
+      "epoch": 2.68,
       "learning_rate": 4.45e-05,
-      "loss": 0.3621,
+      "loss": 0.2508,
       "step": 890
     },
     {
-      "epoch": 0.68,
+      "epoch": 2.71,
       "learning_rate": 4.4999999999999996e-05,
-      "loss": 0.3715,
+      "loss": 0.2247,
       "step": 900
     },
     {
-      "epoch": 0.68,
-      "eval_accuracy": 0.8932164989671486,
-      "eval_loss": 0.33608904480934143,
-      "eval_runtime": 49.5424,
-      "eval_samples_per_second": 605.825,
-      "eval_steps_per_second": 1.191,
+      "epoch": 2.71,
+      "eval_accuracy": 0.9273005930565736,
+      "eval_combined_score": 0.9281516215920332,
+      "eval_f1": 0.9280796047956419,
+      "eval_loss": 0.22854498028755188,
+      "eval_precision": 0.9299256954593439,
+      "eval_recall": 0.9273005930565736,
+      "eval_runtime": 49.2291,
+      "eval_samples_per_second": 609.68,
+      "eval_steps_per_second": 1.198,
       "step": 900
     },
     {
-      "epoch": 0.68,
+      "epoch": 2.74,
       "learning_rate": 4.55e-05,
-      "loss": 0.3337,
+      "loss": 0.2322,
       "step": 910
     },
     {
-      "epoch": 0.69,
+      "epoch": 2.77,
       "learning_rate": 4.6e-05,
-      "loss": 0.3592,
+      "loss": 0.2402,
       "step": 920
     },
     {
-      "epoch": 0.7,
+      "epoch": 2.77,
+      "eval_accuracy": 0.9306323715599387,
+      "eval_combined_score": 0.9309955842160982,
+      "eval_f1": 0.9310096138675644,
+      "eval_loss": 0.23041337728500366,
+      "eval_precision": 0.9317079798769508,
+      "eval_recall": 0.9306323715599387,
+      "eval_runtime": 49.1984,
+      "eval_samples_per_second": 610.061,
+      "eval_steps_per_second": 1.199,
+      "step": 920
+    },
+    {
+      "epoch": 2.8,
       "learning_rate": 4.65e-05,
-      "loss": 0.3845,
+      "loss": 0.2271,
       "step": 930
     },
     {
-      "epoch": 0.71,
+      "epoch": 2.83,
       "learning_rate": 4.7000000000000004e-05,
-      "loss": 0.4175,
+      "loss": 0.2033,
       "step": 940
     },
     {
-      "epoch": 0.71,
-      "learning_rate": 4.75e-05,
-      "loss": 0.4039,
-      "step": 950
+      "epoch": 2.83,
+      "eval_accuracy": 0.9318651296061837,
+      "eval_combined_score": 0.9319597345286359,
+      "eval_f1": 0.9316097686956853,
+      "eval_loss": 0.22282446920871735,
+      "eval_precision": 0.9324989102064907,
+      "eval_recall": 0.9318651296061837,
+      "eval_runtime": 49.2383,
+      "eval_samples_per_second": 609.566,
+      "eval_steps_per_second": 1.198,
+      "step": 940
     },
     {
-      "epoch": 0.71,
-      "eval_accuracy": 0.8966815486106484,
-      "eval_loss": 0.3278920352458954,
-      "eval_runtime": 49.6606,
-      "eval_samples_per_second": 604.383,
-      "eval_steps_per_second": 1.188,
+      "epoch": 2.86,
+      "learning_rate": 4.75e-05,
+      "loss": 0.2296,
       "step": 950
     },
     {
-      "epoch": 0.72,
+      "epoch": 2.89,
       "learning_rate": 4.8e-05,
-      "loss": 0.3474,
+      "loss": 0.2315,
+      "step": 960
+    },
+    {
+      "epoch": 2.89,
+      "eval_accuracy": 0.9270673685613381,
+      "eval_combined_score": 0.928342902973598,
+      "eval_f1": 0.9281145494229922,
+      "eval_loss": 0.22748184204101562,
+      "eval_precision": 0.9311223253487236,
+      "eval_recall": 0.9270673685613381,
+      "eval_runtime": 49.2362,
+      "eval_samples_per_second": 609.592,
+      "eval_steps_per_second": 1.198,
       "step": 960
     },
     {
-      "epoch": 0.73,
+      "epoch": 2.92,
       "learning_rate": 4.85e-05,
-      "loss": 0.3804,
+      "loss": 0.2223,
       "step": 970
     },
     {
-      "epoch": 0.74,
+      "epoch": 2.95,
       "learning_rate": 4.9000000000000005e-05,
-      "loss": 0.3432,
+      "loss": 0.2231,
       "step": 980
     },
     {
-      "epoch": 0.74,
+      "epoch": 2.95,
+      "eval_accuracy": 0.9342973279136403,
+      "eval_combined_score": 0.9345308601987949,
+      "eval_f1": 0.9344884304469883,
+      "eval_loss": 0.21153467893600464,
+      "eval_precision": 0.9350403545209113,
+      "eval_recall": 0.9342973279136403,
+      "eval_runtime": 49.0355,
+      "eval_samples_per_second": 612.087,
+      "eval_steps_per_second": 1.203,
+      "step": 980
+    },
+    {
+      "epoch": 2.98,
       "learning_rate": 4.9500000000000004e-05,
-      "loss": 0.3184,
+      "loss": 0.2282,
       "step": 990
     },
     {
-      "epoch": 0.75,
+      "epoch": 3.01,
       "learning_rate": 5e-05,
-      "loss": 0.369,
+      "loss": 0.2061,
       "step": 1000
     },
     {
-      "epoch": 0.75,
-      "eval_accuracy": 0.9011461318051576,
-      "eval_loss": 0.33548465371131897,
-      "eval_runtime": 49.6573,
-      "eval_samples_per_second": 604.422,
-      "eval_steps_per_second": 1.188,
+      "epoch": 3.01,
+      "eval_accuracy": 0.9355300859598854,
+      "eval_combined_score": 0.9356749651327266,
+      "eval_f1": 0.9355579422022496,
+      "eval_loss": 0.21559903025627136,
+      "eval_precision": 0.936081746408886,
+      "eval_recall": 0.9355300859598854,
+      "eval_runtime": 49.063,
+      "eval_samples_per_second": 611.744,
+      "eval_steps_per_second": 1.203,
       "step": 1000
     }
   ],
   "logging_steps": 10,
-  "max_steps": 3987,
-  "num_train_epochs": 3,
-  "save_steps": 250,
-  "total_flos": 3.3678516289536e+16,
+  "max_steps": 1660,
+  "num_train_epochs": 5,
+  "save_steps": 100,
+  "total_flos": 1.3470775515471053e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:842788c44aca7ccbca2d57d8869e43eb15da3880b4cadb0b091bb46a3e020fcc
+oid sha256:ae98e7de1b05a570517ae68653e3c31b639a52c739d05197601f467f38b01c66
 size 4536