Training in progress, step 2500, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cedf4971a7157fd6bf2d6feebbaf0240f131c23a6aaa72257b8d4af3408f99e4
+oid sha256:6e83191e9de77954f10ec8d943f4fdd1215b54f4fab154b17120e1a226fb4528
 size 5125261

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7686f4fb6dccf6ef68f65a46df2006c3d78f5fe61e8dba208e3cb0272df797c
+oid sha256:0d24372e3cafb13ca988fb39ee3aa6f6339fcfad36ed0d32e5ed1a482b1347d1
 size 2460465086

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ca32586bfda8fd1d34bb7ca8e64a65d045528f0ab228f5d414f71d2c61b4989
+oid sha256:fd154127396e5deb43e241ef5d251275e538c90f80c3d28ad6fc02d2af77d898
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5466d8ea9c57da5074af2cc5d48a8f8664876c6c924307fbf397a610d232702d
+oid sha256:e7e7fc7cce7c11a9cf69d6eb62bc6fa1462ffebea3d00eeb12026f2494eafd55
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.679006338119507,
-  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-2000",
-  "epoch": 0.4406256884776382,
+  "best_metric": 2.5206823348999023,
+  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-2500",
+  "epoch": 0.5507821105970478,
   "eval_steps": 500,
-  "global_step": 2000,
+  "global_step": 2500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -63,13 +63,27 @@
       "eval_samples_per_second": 208.134,
       "eval_steps_per_second": 13.023,
       "step": 2000
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 1.916666666666667e-05,
+      "loss": 2.7296,
+      "step": 2500
+    },
+    {
+      "epoch": 0.55,
+      "eval_loss": 2.5206823348999023,
+      "eval_runtime": 42.8291,
+      "eval_samples_per_second": 211.959,
+      "eval_steps_per_second": 13.262,
+      "step": 2500
     }
   ],
   "logging_steps": 500,
   "max_steps": 60000,
   "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 2025260833701888.0,
+  "total_flos": 2528570877542400.0,
   "trial_name": null,
   "trial_params": null
 }