Training in progress, step 2600, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:625b19eb718436380c2f37fc83ce0b492265d114108abbe7eb0955eb2252a7ca
+oid sha256:ea49c4f6b16598bd78ddbba090e1ab56c7c917b7baaf2ad00a0278dc8d9a629a
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8581a5be5e7f7bc715ad8b96e9a3a0e319b4c12a1b9872f8b846782aec85d420
+oid sha256:7edc328ba69ec88787510e6c7cfbb8faca80990cdd98c151853de2146b487bb0
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3db0aa0b06b80c0bfe7d2cc7b160ec882f2b48b94da908c17282a5c1a88df384
+oid sha256:7376cbf010a9f08e21c74d30e4ee024a7246fb9200b1adb483a435bf9cadb59a
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd37581d93a41fc457d899cacb7a67009a6909f6dbf72b467de55158ef91ac94
+oid sha256:ee2eceee28d7a299780008f817c050eb9e87b0d18fb8d8f19062cda68b3e10f9
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.3950382471084595,
-  "best_model_checkpoint": "./outputs/checkpoint-2500",
-  "epoch": 1.8218579234972676,
+  "best_metric": 1.38124418258667,
+  "best_model_checkpoint": "./outputs/checkpoint-2600",
+  "epoch": 1.8947176684881604,
   "eval_steps": 100,
-  "global_step": 2500,
+  "global_step": 2600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -357,13 +357,27 @@
       "eval_samples_per_second": 46.959,
       "eval_steps_per_second": 5.876,
       "step": 2500
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 2600
+    },
+    {
+      "epoch": 1.89,
+      "eval_loss": 1.38124418258667,
+      "eval_runtime": 133.4827,
+      "eval_samples_per_second": 47.002,
+      "eval_steps_per_second": 5.881,
+      "step": 2600
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.4901801527744102e+17,
+  "total_flos": 1.549783874110341e+17,
   "trial_name": null,
   "trial_params": null
 }