Training in progress, step 17000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c48cffa2aee75c50f2769e5a27e103fb1208a667ce4ef9c1ec217d43f718232
+oid sha256:9175805630a622e73ca492f94b77d5dc0ea902b37d9ae17705b651e16ecd20ef
 size 5125261

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd93eead5db22ed7511a6e711ebe12dceec9da217a07aa6ef5d0eecbaa060da7
+oid sha256:b740d51e796b9fd3c0d8d46f10c04c82c5e93abc46c869fa9a1d904c212f3e7e
 size 2460465086

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfb5c4a03551949fcf48bba2bec36f3e7e3b8146fd809d03227a751e6199f92f
+oid sha256:a985964cfb7e0952c952fbeb01ab98aa97733bf482e8840e1ddd41a5024b841d
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eec233f0702ccbc8f35f54687c96099a02f88403b45595309c416466a5e5b699
+oid sha256:4486a6ff141c5a7c9b6e4200912bd46df4bc7b3ef69c22a34e38b4eb3cc8332f
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.5577419996261597,
-  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-16500",
-  "epoch": 3.635161929940516,
+  "best_metric": 1.544019103050232,
+  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-17000",
+  "epoch": 3.7453183520599254,
   "eval_steps": 500,
-  "global_step": 16500,
+  "global_step": 17000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -469,13 +469,27 @@
       "eval_samples_per_second": 207.908,
       "eval_steps_per_second": 13.009,
       "step": 16500
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 1.4335666666666667e-05,
+      "loss": 1.5219,
+      "step": 17000
+    },
+    {
+      "epoch": 3.75,
+      "eval_loss": 1.544019103050232,
+      "eval_runtime": 43.4005,
+      "eval_samples_per_second": 209.168,
+      "eval_steps_per_second": 13.087,
+      "step": 17000
     }
   ],
   "logging_steps": 500,
   "max_steps": 60000,
   "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 1.6744303013265408e+16,
+  "total_flos": 1.7257839081947136e+16,
   "trial_name": null,
   "trial_params": null
 }