Training in progress, step 2500, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:03c9a04c93f04730cd34b9de4ff0d00f971c0b828bb532c5cce078890208e48f
+oid sha256:625b19eb718436380c2f37fc83ce0b492265d114108abbe7eb0955eb2252a7ca
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e3b42fb9d8c5f41540580d67d8d8f150641f11dc5a9ea914782fcdc5d2c8f67
+oid sha256:8581a5be5e7f7bc715ad8b96e9a3a0e319b4c12a1b9872f8b846782aec85d420
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:376b77f4eda77baae5bb9c56735abd6b9c105452e0922a23656662b4b444f0bc
+oid sha256:3db0aa0b06b80c0bfe7d2cc7b160ec882f2b48b94da908c17282a5c1a88df384
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a78e728336b77d3305afe2734acb1241c76f8f2c33b35460bc07397abc49ba2
+oid sha256:bd37581d93a41fc457d899cacb7a67009a6909f6dbf72b467de55158ef91ac94
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.408768892288208,
-  "best_model_checkpoint": "./outputs/checkpoint-2400",
-  "epoch": 1.7489981785063753,
+  "best_metric": 1.3950382471084595,
+  "best_model_checkpoint": "./outputs/checkpoint-2500",
+  "epoch": 1.8218579234972676,
   "eval_steps": 100,
-  "global_step": 2400,
+  "global_step": 2500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -343,13 +343,27 @@
       "eval_samples_per_second": 46.979,
       "eval_steps_per_second": 5.878,
       "step": 2400
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.0002,
+      "loss": 1.3897,
+      "step": 2500
+    },
+    {
+      "epoch": 1.82,
+      "eval_loss": 1.3950382471084595,
+      "eval_runtime": 133.6055,
+      "eval_samples_per_second": 46.959,
+      "eval_steps_per_second": 5.876,
+      "step": 2500
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.4305055049825485e+17,
+  "total_flos": 1.4901801527744102e+17,
   "trial_name": null,
   "trial_params": null
 }