Training in progress, step 500, checkpoint

checkpoint-500/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:266d1d8179f7989f9b36af575e9b402ceb11422222b40cebe99e87903c573107
+oid sha256:ddbec630d8d4e2ba6cfe8de48d4c8c1f5f2dcf6f31eb064b794a1f3a7a1ec62a
 size 9873829

checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:58865709f250f4ff0501b31b43475fc1203984456d814556216d59d6f4141bde
+oid sha256:32ef827479040b3af3cc83b91a92d9cbaae2a335d6970a9615f303092f3f68ad
 size 42788

checkpoint-500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f7653759f4f1ddd9069725e89aaebdc88edf3a28844142a281b371fb055da08
+oid sha256:0d68fe56cbf15026758f37c8a680b1535157bd18583612558f095fc58469cc5c
 size 14244

checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:529f7b5a081c4e5baab9943dcc97123ba1af7e2a5a8a998441f676f9817f2b58
+oid sha256:cea8b9d0f050de9b969bba9b4cb43afde7581d6291473c7c1b729a06cb63c7a2
 size 1064

checkpoint-500/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.6528732180595398,
-  "best_model_checkpoint": "output/checkpoint-200",
-  "epoch": 0.40108292389451516,
+  "best_metric": 0.6430063247680664,
+  "best_model_checkpoint": "output/checkpoint-50",
+  "epoch": 0.4008417677121956,
   "eval_steps": 50,
   "global_step": 500,
   "is_hyper_param_search": false,
@@ -10,166 +10,166 @@
   "log_history": [
     {
       "epoch": 0.0,
-      "learning_rate": 1.3333333333333332e-06,
-      "loss": 0.785,
+      "learning_rate": 4.444444444444445e-07,
+      "loss": 0.5542,
       "step": 1
     },
     {
       "epoch": 0.04,
-      "learning_rate": 5.9999999999999995e-05,
-      "loss": 0.6547,
+      "learning_rate": 2e-05,
+      "loss": 0.554,
       "step": 50
     },
     {
       "epoch": 0.04,
       "eval_accuracy": 0.6626865671641791,
-      "eval_loss": 0.8300915956497192,
-      "eval_runtime": 61.5689,
-      "eval_samples_per_second": 5.441,
-      "eval_steps_per_second": 1.364,
+      "eval_loss": 0.6430063247680664,
+      "eval_runtime": 61.7692,
+      "eval_samples_per_second": 5.423,
+      "eval_steps_per_second": 1.36,
       "step": 50
     },
     {
       "epoch": 0.08,
-      "learning_rate": 0.00012666666666666666,
-      "loss": 0.6654,
+      "learning_rate": 4.222222222222222e-05,
+      "loss": 0.5957,
       "step": 100
     },
     {
       "epoch": 0.08,
-      "eval_accuracy": 0.6895522388059702,
-      "eval_loss": 0.6663276553153992,
-      "eval_runtime": 56.4633,
-      "eval_samples_per_second": 5.933,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6686567164179105,
+      "eval_loss": 0.6705919504165649,
+      "eval_runtime": 60.4388,
+      "eval_samples_per_second": 5.543,
+      "eval_steps_per_second": 1.39,
       "step": 100
     },
     {
       "epoch": 0.12,
-      "learning_rate": 0.00019333333333333333,
-      "loss": 0.6491,
+      "learning_rate": 6.444444444444446e-05,
+      "loss": 0.5668,
       "step": 150
     },
     {
       "epoch": 0.12,
-      "eval_accuracy": 0.6537313432835821,
-      "eval_loss": 0.9172552824020386,
-      "eval_runtime": 56.4705,
-      "eval_samples_per_second": 5.932,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6805970149253732,
+      "eval_loss": 0.7224913835525513,
+      "eval_runtime": 56.4882,
+      "eval_samples_per_second": 5.93,
+      "eval_steps_per_second": 1.487,
       "step": 150
     },
     {
       "epoch": 0.16,
-      "learning_rate": 0.00026,
-      "loss": 0.641,
+      "learning_rate": 8.666666666666667e-05,
+      "loss": 0.5626,
       "step": 200
     },
     {
       "epoch": 0.16,
-      "eval_accuracy": 0.6865671641791045,
-      "eval_loss": 0.6528732180595398,
-      "eval_runtime": 56.4485,
-      "eval_samples_per_second": 5.935,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6746268656716418,
+      "eval_loss": 0.7115103006362915,
+      "eval_runtime": 56.506,
+      "eval_samples_per_second": 5.929,
+      "eval_steps_per_second": 1.487,
       "step": 200
     },
     {
       "epoch": 0.2,
-      "learning_rate": 0.0002941234084231146,
-      "loss": 0.6521,
+      "learning_rate": 9.80430528375734e-05,
+      "loss": 0.5604,
       "step": 250
     },
     {
       "epoch": 0.2,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.8197052478790283,
-      "eval_runtime": 56.4577,
-      "eval_samples_per_second": 5.934,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6746268656716418,
+      "eval_loss": 0.700947105884552,
+      "eval_runtime": 56.4927,
+      "eval_samples_per_second": 5.93,
+      "eval_steps_per_second": 1.487,
       "step": 250
     },
     {
       "epoch": 0.24,
-      "learning_rate": 0.00027943192948090106,
-      "loss": 0.638,
+      "learning_rate": 9.315068493150684e-05,
+      "loss": 0.5675,
       "step": 300
     },
     {
       "epoch": 0.24,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.6961421370506287,
-      "eval_runtime": 56.4506,
-      "eval_samples_per_second": 5.934,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6597014925373135,
+      "eval_loss": 0.662155032157898,
+      "eval_runtime": 56.4903,
+      "eval_samples_per_second": 5.93,
+      "eval_steps_per_second": 1.487,
       "step": 300
     },
     {
       "epoch": 0.28,
-      "learning_rate": 0.00026474045053868754,
-      "loss": 0.6857,
+      "learning_rate": 8.825831702544032e-05,
+      "loss": 0.5725,
       "step": 350
     },
     {
       "epoch": 0.28,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.7876304388046265,
-      "eval_runtime": 56.4651,
-      "eval_samples_per_second": 5.933,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6567164179104478,
+      "eval_loss": 0.6746364831924438,
+      "eval_runtime": 56.482,
+      "eval_samples_per_second": 5.931,
+      "eval_steps_per_second": 1.487,
       "step": 350
     },
     {
       "epoch": 0.32,
-      "learning_rate": 0.000250048971596474,
-      "loss": 0.6469,
+      "learning_rate": 8.336594911937378e-05,
+      "loss": 0.5884,
       "step": 400
     },
     {
       "epoch": 0.32,
-      "eval_accuracy": 0.6626865671641791,
-      "eval_loss": 0.7487305402755737,
-      "eval_runtime": 56.4489,
-      "eval_samples_per_second": 5.935,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6656716417910448,
+      "eval_loss": 0.7047849893569946,
+      "eval_runtime": 56.4873,
+      "eval_samples_per_second": 5.931,
+      "eval_steps_per_second": 1.487,
       "step": 400
     },
     {
       "epoch": 0.36,
-      "learning_rate": 0.00023535749265426052,
-      "loss": 0.6813,
+      "learning_rate": 7.847358121330725e-05,
+      "loss": 0.5567,
       "step": 450
     },
     {
       "epoch": 0.36,
-      "eval_accuracy": 0.6626865671641791,
-      "eval_loss": 0.8447557091712952,
-      "eval_runtime": 56.4428,
-      "eval_samples_per_second": 5.935,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6477611940298508,
+      "eval_loss": 0.6832321286201477,
+      "eval_runtime": 56.4814,
+      "eval_samples_per_second": 5.931,
+      "eval_steps_per_second": 1.487,
       "step": 450
     },
     {
       "epoch": 0.4,
-      "learning_rate": 0.000220666013712047,
-      "loss": 0.645,
+      "learning_rate": 7.35812133072407e-05,
+      "loss": 0.5723,
       "step": 500
     },
     {
       "epoch": 0.4,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.6624641418457031,
-      "eval_runtime": 56.4531,
-      "eval_samples_per_second": 5.934,
+      "eval_accuracy": 0.6955223880597015,
+      "eval_loss": 0.7437570691108704,
+      "eval_runtime": 56.4677,
+      "eval_samples_per_second": 5.933,
       "eval_steps_per_second": 1.488,
       "step": 500
     }
   ],
   "logging_steps": 50,
-  "max_steps": 1246,
+  "max_steps": 1247,
   "num_train_epochs": 1,
-  "save_steps": 100,
-  "total_flos": 1.86852300708864e+17,
+  "save_steps": 50,
+  "total_flos": 1.8605381592588288e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e32e120c6d65576e6fe76fdf63d907877e27469586cc746d8cd6a71d07635ea9
+oid sha256:87bd30bd4b21b4a0c66f1fd545858ef6e2d33b5984a8d292e79011acc3affd6b
 size 4472