Training in progress, epoch 3

checkpoint-540/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e08513b86007464454b0e6f18328f1276f7c2d917f8f3cb3bb5234e9f8f7895
+oid sha256:e68a10c404c5d2605b9d33fbc45deff231a03f06f6a4eb191406ec035bbb6f49
 size 1115518581

checkpoint-540/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b3860f71d58616e27ca1900afe52e2e411611c1648789afb5252cd57b42fc6e7
+oid sha256:2d1c0097f49c1da6253017544d8b7e2be9ad69411aef2cea8d2f938d4d27d075
 size 557982265

checkpoint-540/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:620db0493c00c974636df70310a208d6729ef8f4b083036c4ff9d8876742822e
+oid sha256:b719e1cf279d8f44ff232dfbca518fa1d7c2493350504703531dcf21c6259177
 size 623

checkpoint-540/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.1801992505788803,
+  "best_metric": 0.19929151237010956,
   "best_model_checkpoint": "outputs/bart-base-detox/checkpoint-405",
   "epoch": 3.9953917050691246,
   "global_step": 540,
@@ -9,58 +9,58 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 9e-05,
-      "loss": 0.3171,
+      "learning_rate": 9e-06,
+      "loss": 0.5633,
       "step": 135
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.20179320871829987,
-      "eval_runtime": 11.3072,
-      "eval_samples_per_second": 274.16,
-      "eval_steps_per_second": 34.314,
+      "eval_loss": 0.25235414505004883,
+      "eval_runtime": 12.2006,
+      "eval_samples_per_second": 254.085,
+      "eval_steps_per_second": 31.802,
       "step": 135
     },
     {
       "epoch": 2.0,
-      "learning_rate": 8e-05,
-      "loss": 0.1852,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.2589,
       "step": 270
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.18728719651699066,
-      "eval_runtime": 11.1563,
-      "eval_samples_per_second": 277.869,
-      "eval_steps_per_second": 34.778,
+      "eval_loss": 0.21927106380462646,
+      "eval_runtime": 11.2792,
+      "eval_samples_per_second": 274.842,
+      "eval_steps_per_second": 34.4,
       "step": 270
     },
     {
       "epoch": 3.0,
-      "learning_rate": 7e-05,
-      "loss": 0.1559,
+      "learning_rate": 7e-06,
+      "loss": 0.2307,
       "step": 405
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.1801992505788803,
-      "eval_runtime": 11.3199,
-      "eval_samples_per_second": 273.853,
-      "eval_steps_per_second": 34.276,
+      "eval_loss": 0.19929151237010956,
+      "eval_runtime": 11.2313,
+      "eval_samples_per_second": 276.016,
+      "eval_steps_per_second": 34.546,
       "step": 405
     },
     {
       "epoch": 4.0,
-      "learning_rate": 6e-05,
-      "loss": 0.1356,
+      "learning_rate": 6e-06,
+      "loss": 0.2171,
       "step": 540
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.18711909651756287,
-      "eval_runtime": 11.3935,
-      "eval_samples_per_second": 272.085,
-      "eval_steps_per_second": 34.054,
+      "eval_loss": 0.20016591250896454,
+      "eval_runtime": 11.0923,
+      "eval_samples_per_second": 279.473,
+      "eval_steps_per_second": 34.979,
       "step": 540
     }
   ],

checkpoint-540/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6779c0d7bfc5cc8b7f502411db449902ffdb3fd6d5cc4758f90defaa037cdfa
+oid sha256:cd8019ba81e4d8cd7e878f72b37c14bf5f28f7f6b5acd2cf343bb5ba08c77b6a
 size 3183

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61ada5f16a59af821e82557d0cc1154b9a8317ada4b526eedac4e607d1c6f078
+oid sha256:2d1c0097f49c1da6253017544d8b7e2be9ad69411aef2cea8d2f938d4d27d075
 size 557982265