Training in progress, step 3500

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:118f11e0d84ce3412f0281c67abd06b029f3d4d07a72438ba7182ce0f59d4e32
+oid sha256:6413b446cd4cf9555dcc70e093250592683c054ac330b802b68b79522db341d1
 size 484015877

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14065493d9c0cd89909a4479c6adb4522d23ff4fbcc1c3c0ac202a7603403dc3
+oid sha256:b573b82c4722ab456322756912925fd8af33c728c413d87733bb7b99f08ae499
 size 242014297

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2cba09060d4e5c0be85a1d41e58c89ab09d39f09122d03e1bc8222b013265c96
+oid sha256:40b5adfb915bd56ca45aa103901c57ec3fa114d95a597006e52879a68aa6590a
 size 14575

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d821d43f1f40a9df79baae3da2b634f271a408f300c294f73c823b49cd908ce5
+oid sha256:b8910ac4d4e368661aaca00b9d59b975c6b600a74099129e4a39e35e40ad9e9d
 size 557

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c87425c169975acef42a6cd137204ef2ac5b636056155eafc97d710301d2b0c
+oid sha256:14a52e7813087beccf319e8483ed659abb6bbb1fda8eb608a9bf0241413c19ba
 size 627

last-checkpoint/trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.5571591325212851,
-  "best_model_checkpoint": "results/checkpoint-3000",
-  "epoch": 0.6,
-  "global_step": 3000,
+  "best_metric": 0.5598750842730023,
+  "best_model_checkpoint": "results/checkpoint-3500",
+  "epoch": 0.7,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -126,11 +126,31 @@
       "eval_samples_per_second": 35.119,
       "eval_steps_per_second": 1.756,
       "step": 3000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 8.073684210526316e-05,
+      "loss": 0.6923,
+      "step": 3500
+    },
+    {
+      "epoch": 0.7,
+      "eval_bleu": 0.1269094917224047,
+      "eval_exact_match": 0.156,
+      "eval_loss": 0.8161399960517883,
+      "eval_rouge1": 0.5598750842730023,
+      "eval_rouge2": 0.38392351740583797,
+      "eval_rougeL": 0.5409288136858783,
+      "eval_rougeLsum": 0.5484962105303921,
+      "eval_runtime": 55.0573,
+      "eval_samples_per_second": 36.326,
+      "eval_steps_per_second": 1.816,
+      "step": 3500
     }
   ],
   "max_steps": 15000,
   "num_train_epochs": 3,
-  "total_flos": 7695000865996800.0,
+  "total_flos": 8985722848542720.0,
   "trial_name": null,
   "trial_params": null
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14065493d9c0cd89909a4479c6adb4522d23ff4fbcc1c3c0ac202a7603403dc3
+oid sha256:b573b82c4722ab456322756912925fd8af33c728c413d87733bb7b99f08ae499
 size 242014297