Training in progress, step 36000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f089ca63a8be28b0b3efc08204d1937c824424cb84f62f2a98d5cad74de540f2
+oid sha256:0823af30b457b21d03622b0e5ec607830ad3eb4931531ef4e66fdd7b5d848945
 size 5125261

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e070de007c58a4d3419fc03b6d8b8681766ee17017bba7a112f9373f74262005
+oid sha256:e560e12102cb30c8844e9201276e5e4411a817ddf7088c208b2d192b75a8786e
 size 2460465086

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bb0821d7b1115e7bd56afb45eae715aab192afff0e226a97d34c7c138215a92
+oid sha256:26c038200f4f8a74c11a8785b8e167d9df1bbab710a7e672e0abeafc1b98201d
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c32a89b05e9441bfcdff8d54b4a37933812368e5f991d0aa3b7462ffb29ad720
+oid sha256:2b506a29889e6812ebf3d570fcad99df58085bc924b25685203b74a2b743ec3a
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.3616106510162354,
-  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-35500",
-  "epoch": 7.8211059704780785,
+  "best_metric": 1.357163429260254,
+  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-36000",
+  "epoch": 7.931262392597488,
   "eval_steps": 500,
-  "global_step": 35500,
+  "global_step": 36000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1001,13 +1001,27 @@
       "eval_samples_per_second": 199.792,
       "eval_steps_per_second": 12.501,
       "step": 35500
+    },
+    {
+      "epoch": 7.93,
+      "learning_rate": 8.004e-06,
+      "loss": 1.2261,
+      "step": 36000
+    },
+    {
+      "epoch": 7.93,
+      "eval_loss": 1.357163429260254,
+      "eval_runtime": 45.4325,
+      "eval_samples_per_second": 199.813,
+      "eval_steps_per_second": 12.502,
+      "step": 36000
     }
   ],
   "logging_steps": 500,
   "max_steps": 60000,
   "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 3.599083303875379e+16,
+  "total_flos": 3.650436910743552e+16,
   "trial_name": null,
   "trial_params": null
 }