Training in progress, step 2500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6af466bfad7b84d3f92fdc82d61cd05694f71ff9eecc943c6b5592bb1d3e564d
+oid sha256:878a35f34cd75f45d403c328a94efa5875841d2b2917a026661461fe9a326377
 size 1625422896

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ff6601686031b7765f34ea912c7bbc2327ba935664b6facb5fb29fd2b6e467a
+oid sha256:0f0a69880690b268acd014dbb76f6b9dbeac3ab805b96ce286bbc304abcabbb0
 size 3250759392

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8de8b3ca8b0346edfe159c193ca900a7ac29bd3f6c2ea3b1718b858590ac10e
+oid sha256:19942036ae732bf0600dafcf1c4422b97c2fe83799394bc0c6ff948aabb78ad9
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f70a28f809f2e99aba9c326954681b23221caf6b22c57b9fb6e81d614c34b1f9
+oid sha256:5d09b05bd39538ef18816113a9132fa750a22f57bb959571defdc79d084dac52
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d51a482c6985f87480f189bb302d591d56fe7879c8ecc114a67e898aa53b82ca
+oid sha256:16c7d20f5d8e2a0e338070c52ebbebc70e1adc04ba4e4f2f006768811444edf6
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 2000,
   "best_metric": 50.8,
   "best_model_checkpoint": "./bart-email-multitask-updated/checkpoint-2000",
-  "epoch": 0.6904289289721239,
+  "epoch": 0.8630361612151549,
   "eval_steps": 500,
-  "global_step": 2000,
+  "global_step": 2500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -340,6 +340,89 @@
       "eval_samples_per_second": 1.095,
       "eval_steps_per_second": 0.274,
       "step": 2000
+    },
+    {
+      "epoch": 0.707689652196427,
+      "grad_norm": 5.292072772979736,
+      "learning_rate": 3.3473364318913127e-06,
+      "loss": 0.4216,
+      "step": 2050
+    },
+    {
+      "epoch": 0.7249503754207302,
+      "grad_norm": 2.9521191120147705,
+      "learning_rate": 3.3026456918126563e-06,
+      "loss": 0.4332,
+      "step": 2100
+    },
+    {
+      "epoch": 0.7422110986450332,
+      "grad_norm": 4.375521183013916,
+      "learning_rate": 3.2579549517340008e-06,
+      "loss": 0.4392,
+      "step": 2150
+    },
+    {
+      "epoch": 0.7594718218693364,
+      "grad_norm": 3.5504395961761475,
+      "learning_rate": 3.2132642116553453e-06,
+      "loss": 0.432,
+      "step": 2200
+    },
+    {
+      "epoch": 0.7767325450936394,
+      "grad_norm": 3.7240469455718994,
+      "learning_rate": 3.1685734715766897e-06,
+      "loss": 0.4278,
+      "step": 2250
+    },
+    {
+      "epoch": 0.7939932683179425,
+      "grad_norm": 2.717268943786621,
+      "learning_rate": 3.123882731498034e-06,
+      "loss": 0.4145,
+      "step": 2300
+    },
+    {
+      "epoch": 0.8112539915422456,
+      "grad_norm": 2.7192165851593018,
+      "learning_rate": 3.079191991419378e-06,
+      "loss": 0.4454,
+      "step": 2350
+    },
+    {
+      "epoch": 0.8285147147665487,
+      "grad_norm": 4.111413955688477,
+      "learning_rate": 3.0345012513407223e-06,
+      "loss": 0.4188,
+      "step": 2400
+    },
+    {
+      "epoch": 0.8457754379908519,
+      "grad_norm": 3.396127223968506,
+      "learning_rate": 2.9898105112620668e-06,
+      "loss": 0.4349,
+      "step": 2450
+    },
+    {
+      "epoch": 0.8630361612151549,
+      "grad_norm": 2.503493309020996,
+      "learning_rate": 2.9451197711834112e-06,
+      "loss": 0.4133,
+      "step": 2500
+    },
+    {
+      "epoch": 0.8630361612151549,
+      "eval_gen_len": 54.88080279232112,
+      "eval_loss": 0.5358440279960632,
+      "eval_rouge1": 50.67,
+      "eval_rouge2": 36.36,
+      "eval_rougeL": 44.02,
+      "eval_rougeLsum": 44.56,
+      "eval_runtime": 5344.428,
+      "eval_samples_per_second": 1.072,
+      "eval_steps_per_second": 0.268,
+      "step": 2500
     }
   ],
   "logging_steps": 50,
@@ -354,7 +437,7 @@
         "early_stopping_threshold": 0.0
       },
       "attributes": {
-        "early_stopping_patience_counter": 0
+        "early_stopping_patience_counter": 1
       }
     },
     "TrainerControl": {
@@ -368,7 +451,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.8670127079424e+16,
+  "total_flos": 8.583765884928e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null