Training in progress, step 20000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f38d74263534693877152baa21dc59d875f3378c0490ae05ef33eb6dc5f8642
+oid sha256:04b41db138ee8e198f9e2a7b0efd7791bc14180db542c590d7ce8a9f5babd6ef
 size 5125261

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a5b66ab356d63c0d08ccf26aeb3eb34e988153f75e8d9b4862476999dfba889
+oid sha256:870a2832da25c2f26b5846c3556adfe33aff26954baa3852c79ff72069a23fc8
 size 2460465086

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fe1c037a79f5e1ab883af9fdcdcfca89747b7f4e8b7eb549eb91d0066611ac7
+oid sha256:5f6c3cb8c70b6e982e63c05f3abfa9ad5033f15f0d861edf82a1c29cf22cd8ef
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:970ff72f9cb5a846ce88a5ea203dc3b82f596d6951a3c394abb92f4ff9fb3a2d
+oid sha256:09a8b2cb77693ed976ba1ef3c7f415f6782867c443fd911c21fd3a0d2839b5c1
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.503504753112793,
-  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-19500",
-  "epoch": 4.296100462656973,
+  "best_metric": 1.495379090309143,
+  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-20000",
+  "epoch": 4.406256884776383,
   "eval_steps": 500,
-  "global_step": 19500,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -553,13 +553,27 @@
       "eval_samples_per_second": 209.952,
       "eval_steps_per_second": 13.136,
       "step": 19500
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 1.3335666666666667e-05,
+      "loss": 1.4563,
+      "step": 20000
+    },
+    {
+      "epoch": 4.41,
+      "eval_loss": 1.495379090309143,
+      "eval_runtime": 43.6248,
+      "eval_samples_per_second": 208.093,
+      "eval_steps_per_second": 13.02,
+      "step": 20000
     }
   ],
   "logging_steps": 500,
   "max_steps": 60000,
   "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 1.9783227024605184e+16,
+  "total_flos": 2.029327540931789e+16,
   "trial_name": null,
   "trial_params": null
 }