Training in progress, step 21000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:245c98e0d04ea116d9f4c8104f7abf1428f7d3663a58acc3f00b215d65c7844d
+oid sha256:0532922c0755883c1f75431e58cd197c55e739151b005def28c2978caefe0b99
 size 5125261

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aea00709e462e96fe448f991df8b661b341798c491fc7874c081b3571381daec
+oid sha256:9bf8d8456febe4ee85e1c4ec003a2db58283eafbfe284d5446916b79f3e7a0ac
 size 2460465086

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd34a70a57c7edd7b2d61dcf32a825b851d6393642456c0bbddfaa5eef14b2a6
+oid sha256:9f600d0753ed00fe58785f22863b5c8f5c8aab579af93a924b04ef738081a16d
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f996421c6fa5bd9101a9ea30b4085e3f14ac40ecc62c043ff0d36b92d20c4015
+oid sha256:7633c094444bb7e56a9f9b4293e2cee3115c6d985113b21c584ad616baf8d5ca
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.4834085702896118,
-  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-20500",
-  "epoch": 4.516413306895792,
+  "best_metric": 1.4811300039291382,
+  "best_model_checkpoint": "smilemikan/nllb-finetuned-jpn-to-ain-2/checkpoint-21000",
+  "epoch": 4.626569729015202,
   "eval_steps": 500,
-  "global_step": 20500,
+  "global_step": 21000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -581,13 +581,27 @@
       "eval_samples_per_second": 206.583,
       "eval_steps_per_second": 12.926,
       "step": 20500
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 1.3002333333333334e-05,
+      "loss": 1.4284,
+      "step": 21000
+    },
+    {
+      "epoch": 4.63,
+      "eval_loss": 1.4811300039291382,
+      "eval_runtime": 43.2207,
+      "eval_samples_per_second": 210.038,
+      "eval_steps_per_second": 13.142,
+      "step": 21000
     }
   ],
   "logging_steps": 500,
   "max_steps": 60000,
   "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 2.0803933292199936e+16,
+  "total_flos": 2.130825916632269e+16,
   "trial_name": null,
   "trial_params": null
 }