Training in progress, step 4000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc2fa15a5e9fc5174a2f2db460995507bf10f8bbf53cd1a2df5dc209c5ea3b39
+oid sha256:738fe809067ec27fe3c901868d7de44ce62aeb2f32711ce135f92d17c0b810a8
 size 5125261

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b5ac68c94586e230c3959acb54f6801fdcc53cbf82147c8780a12f19b5068e1
+oid sha256:74322843b1da9a146c9851ea836c1bebd2d834e2c66c879e38e21abc44eb38ca
 size 2460465086

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a287a7b84031c05a48ea7c5333d8eebf366f57167dc20774be9b71117c6df912
+oid sha256:1255a4808803555132c31bf2257b79123162df6c92908f3a84b09e150d37cd61
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc99e244e20972590fe93655e8af347e514b6b879b75566c51444528a67c5dee
+oid sha256:5a2e5ff60d578727c0bc9cc9ae5e543398968c07cf3ea974c025608e21631cd0
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.30385422706604,
-  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-3500",
-  "epoch": 0.771094954835867,
+  "best_metric": 2.2251899242401123,
+  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-4000",
+  "epoch": 0.8812513769552764,
   "eval_steps": 500,
-  "global_step": 3500,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -105,13 +105,27 @@
       "eval_samples_per_second": 203.096,
       "eval_steps_per_second": 12.707,
       "step": 3500
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.8667000000000003e-05,
+      "loss": 2.3848,
+      "step": 4000
+    },
+    {
+      "epoch": 0.88,
+      "eval_loss": 2.2251899242401123,
+      "eval_runtime": 42.8042,
+      "eval_samples_per_second": 212.082,
+      "eval_steps_per_second": 13.27,
+      "step": 4000
     }
   ],
   "logging_steps": 500,
   "max_steps": 60000,
   "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 3540439421681664.0,
+  "total_flos": 4045442515992576.0,
   "trial_name": null,
   "trial_params": null
 }