Training in progress, epoch 0, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14aa548ee07ab31052157712a9e0da885e6a07b960f6e223e74213cb1443e6fa
+oid sha256:909bf20fd33dfc2259e4df422f1bda32e44a0b9bf3d95c7a7e236b3464d3eff8
 size 1219380555

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:550d262c1ae5d8f37d1984f58337c1c2fc94034c0ddfc220eff610ba015afb5f
+oid sha256:8f3f8f5541517ae2c219789435c043a408058ae7f5c6d1ac4aec5f2156a0019a
 size 652423827

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c90dc5c3c5d48b21df844d28ac5f4dbdc02f7f7e7446ee48fcb6f3327de872fd
+oid sha256:baa26f0a90138d1c3ea988ee3460c5f3e821f7b8e23c1f705aa1f04cf0e90744
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:030ce4e896dde4e5df8ae1c595b2e24f4e2c44ea8055948beb08188b9a047014
+oid sha256:ba889a9edbbc48a836dabf466c2ef5e898237108147ba44abda4201a588a35a8
 size 1465

last-checkpoint/trainer_state.json

@@ -1,49 +1,76 @@
 {
-  "best_metric": 1.3624887493510058,
-  "best_model_checkpoint": "./routed_longformer_docmt/checkpoint-36",
-  "epoch": 2.88,
+  "best_metric": 19.45676397235889,
+  "best_model_checkpoint": "./routed_longformer_docmt/checkpoint-786",
+  "epoch": 0.9993642720915448,
   "eval_steps": 500,
-  "global_step": 36,
+  "global_step": 786,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.96,
-      "eval_bleu": 0.7649978858780119,
-      "eval_chrf": 15.452986698532497,
-      "eval_comet": 0.24576674342155458,
-      "eval_loss": 3.4590535163879395,
-      "eval_runtime": 31.0399,
-      "eval_samples_per_second": 1.611,
-      "eval_steps_per_second": 0.226,
-      "step": 12
+      "epoch": 0.12714558169103624,
+      "grad_norm": 2.7216010093688965,
+      "learning_rate": 2.875318066157761e-05,
+      "loss": 3.5051,
+      "step": 100
     },
     {
-      "epoch": 2.0,
-      "eval_bleu": 1.2647603736034783,
-      "eval_chrf": 16.316051179572895,
-      "eval_comet": 0.23885925143957137,
-      "eval_loss": 3.3582441806793213,
-      "eval_runtime": 31.8746,
-      "eval_samples_per_second": 1.569,
-      "eval_steps_per_second": 0.22,
-      "step": 25
+      "epoch": 0.25429116338207247,
+      "grad_norm": 2.2804791927337646,
+      "learning_rate": 2.7480916030534352e-05,
+      "loss": 2.6407,
+      "step": 200
     },
     {
-      "epoch": 2.88,
-      "eval_bleu": 1.3624887493510058,
-      "eval_chrf": 16.339390714055682,
-      "eval_comet": 0.23527576833963393,
-      "eval_loss": 3.3164820671081543,
-      "eval_runtime": 31.9738,
-      "eval_samples_per_second": 1.564,
-      "eval_steps_per_second": 0.219,
-      "step": 36
+      "epoch": 0.3814367450731087,
+      "grad_norm": 2.807426691055298,
+      "learning_rate": 2.6208651399491097e-05,
+      "loss": 2.3397,
+      "step": 300
+    },
+    {
+      "epoch": 0.5085823267641449,
+      "grad_norm": 1.993668556213379,
+      "learning_rate": 2.4936386768447838e-05,
+      "loss": 2.1398,
+      "step": 400
+    },
+    {
+      "epoch": 0.6357279084551812,
+      "grad_norm": 2.3265788555145264,
+      "learning_rate": 2.3664122137404583e-05,
+      "loss": 2.0065,
+      "step": 500
+    },
+    {
+      "epoch": 0.7628734901462174,
+      "grad_norm": 1.7298423051834106,
+      "learning_rate": 2.2391857506361324e-05,
+      "loss": 1.8887,
+      "step": 600
+    },
+    {
+      "epoch": 0.8900190718372537,
+      "grad_norm": 2.18813419342041,
+      "learning_rate": 2.1119592875318066e-05,
+      "loss": 1.8264,
+      "step": 700
+    },
+    {
+      "epoch": 0.9993642720915448,
+      "eval_bleu": 19.45676397235889,
+      "eval_chrf": 53.07001767638457,
+      "eval_comet": 0.4521310674039932,
+      "eval_loss": 1.5774508714675903,
+      "eval_runtime": 710.5738,
+      "eval_samples_per_second": 1.967,
+      "eval_steps_per_second": 0.246,
+      "step": 786
     }
   ],
   "logging_steps": 100,
-  "max_steps": 36,
+  "max_steps": 2358,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -54,7 +81,7 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf242734023e796808cd7ba63988f57ad6a54ba45632e7e45726f5e59fb4ed9c
+oid sha256:e2f0c57289e38a06b3be61d14dc67c7ac6473c96a88fcc15db5bf382d46f7a10
 size 5713