Training in progress, step 3500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:067b8a7c716303526670dfb8bc56fae03cdbbf29ed37d4991fe64c928e0bdbd0
+oid sha256:e970ebd5a85878e24d4e82408f1544b44510ad867e6e14413abc73b59a6e2b87
 size 1625422896

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:981f8568e18df91cfeab6eec5f61217695b23396cd15d999934d53fd8ca2d195
+oid sha256:9015c6e097a9d0d0b930f42989c855220716dbcdd318d26faaab572c32352885
 size 3250759392

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca1292d0949d9001f354d7b1957f5d7fb3c5eda5b3075a3cdda1634add393adf
+oid sha256:29088cd044384ff39ead7e6da89a896d2c14ec65f5b780b791b52bf8b2923887
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b52c0d5230cbe4dff3dcbc875016ea8649032c17265bd3fa32928b7d60fd030d
+oid sha256:917c700dc3c784b68d992933d13de0aadf8e6b69ea6214efe3a71da31083190f
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a77e0e803bec330fe1d60d35d35c94f06878c2de756c5dca3c6a6f9c076ae62e
+oid sha256:6cf9a735ef029846b4b68a529e8290b963a10b9c42b981859787d6acbeb3970d
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 3000,
   "best_metric": 51.14,
   "best_model_checkpoint": "./bart-email-multitask-updated/checkpoint-3000",
-  "epoch": 1.0355570898420643,
+  "epoch": 1.2081643220850953,
   "eval_steps": 500,
-  "global_step": 3000,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -506,6 +506,89 @@
       "eval_samples_per_second": 1.099,
       "eval_steps_per_second": 0.275,
       "step": 3000
+    },
+    {
+      "epoch": 1.0528178130663675,
+      "grad_norm": 2.5936224460601807,
+      "learning_rate": 2.4535216303181983e-06,
+      "loss": 0.3795,
+      "step": 3050
+    },
+    {
+      "epoch": 1.0700785362906706,
+      "grad_norm": 2.727391481399536,
+      "learning_rate": 2.4088308902395428e-06,
+      "loss": 0.3797,
+      "step": 3100
+    },
+    {
+      "epoch": 1.0873392595149736,
+      "grad_norm": 2.1693854331970215,
+      "learning_rate": 2.364140150160887e-06,
+      "loss": 0.3809,
+      "step": 3150
+    },
+    {
+      "epoch": 1.1045999827392767,
+      "grad_norm": 3.9267098903656006,
+      "learning_rate": 2.3194494100822313e-06,
+      "loss": 0.3816,
+      "step": 3200
+    },
+    {
+      "epoch": 1.12186070596358,
+      "grad_norm": 3.905186176300049,
+      "learning_rate": 2.2747586700035753e-06,
+      "loss": 0.3852,
+      "step": 3250
+    },
+    {
+      "epoch": 1.139121429187883,
+      "grad_norm": 3.1914596557617188,
+      "learning_rate": 2.23006792992492e-06,
+      "loss": 0.3837,
+      "step": 3300
+    },
+    {
+      "epoch": 1.156382152412186,
+      "grad_norm": 3.2150516510009766,
+      "learning_rate": 2.185377189846264e-06,
+      "loss": 0.3916,
+      "step": 3350
+    },
+    {
+      "epoch": 1.1736428756364892,
+      "grad_norm": 3.4676005840301514,
+      "learning_rate": 2.1406864497676083e-06,
+      "loss": 0.3889,
+      "step": 3400
+    },
+    {
+      "epoch": 1.1909035988607923,
+      "grad_norm": 3.015994071960449,
+      "learning_rate": 2.095995709688953e-06,
+      "loss": 0.3901,
+      "step": 3450
+    },
+    {
+      "epoch": 1.2081643220850953,
+      "grad_norm": 3.504163980484009,
+      "learning_rate": 2.051304969610297e-06,
+      "loss": 0.3878,
+      "step": 3500
+    },
+    {
+      "epoch": 1.2081643220850953,
+      "eval_gen_len": 54.42425828970332,
+      "eval_loss": 0.5358798503875732,
+      "eval_rouge1": 50.77,
+      "eval_rouge2": 36.52,
+      "eval_rougeL": 44.17,
+      "eval_rougeLsum": 44.73,
+      "eval_runtime": 5360.6563,
+      "eval_samples_per_second": 1.069,
+      "eval_steps_per_second": 0.267,
+      "step": 3500
     }
   ],
   "logging_steps": 50,
@@ -520,7 +603,7 @@
         "early_stopping_threshold": 0.0
       },
       "attributes": {
-        "early_stopping_patience_counter": 0
+        "early_stopping_patience_counter": 1
       }
     },
     "TrainerControl": {
@@ -534,7 +617,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.029923149703086e+17,
+  "total_flos": 1.201598467401646e+17,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null