Training in progress, step 18500, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1535202f47ea8847be7177abc3b113de060034c4d967d0a1efac4c2724c2068
+oid sha256:ee8d14031344632f9a83d65cbeb334fd222e2206788562f1d0ef7f92f6755dc5
 size 5125261

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2567983426d814eb583e058844d7c3c949c458d3e2ba5ab187de22f62974f1d
+oid sha256:9c9b66246e212762c9bce54efe452a5660f3a27fc5f3db25a15258d1b546cd7e
 size 2460465086

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9ff9a0beb47a5c694497e062f42a55477bbdb638a8059fee9b186d6bcf6111c
+oid sha256:b3a5307f1b9f3a753ffca16a6487033857c954277ede0e43f06c48d6e6e506a1
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9383bd703f2e2b314358fbcfff4794afb50f09dbec39a256e4e3fe424b7952e7
+oid sha256:54e83d3d1c149aaff4f610278a355ef4e3e24cd2d1673a5e4041c0a10c815d4e
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.5262504816055298,
-  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-18000",
-  "epoch": 3.965631196298744,
+  "best_metric": 1.519882082939148,
+  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-18500",
+  "epoch": 4.075787618418154,
   "eval_steps": 500,
-  "global_step": 18000,
+  "global_step": 18500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -511,13 +511,27 @@
       "eval_samples_per_second": 209.477,
       "eval_steps_per_second": 13.107,
       "step": 18000
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 1.3835666666666667e-05,
+      "loss": 1.4771,
+      "step": 18500
+    },
+    {
+      "epoch": 4.08,
+      "eval_loss": 1.519882082939148,
+      "eval_runtime": 43.8272,
+      "eval_samples_per_second": 207.132,
+      "eval_steps_per_second": 12.96,
+      "step": 18500
     }
   ],
   "logging_steps": 500,
   "max_steps": 60000,
   "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 1.827475291648819e+16,
+  "total_flos": 1.8773830334152704e+16,
   "trial_name": null,
   "trial_params": null
 }