Training in progress, step 1500, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:287cdc0db0836385fc8b23617ba8d112ec3a39080cdfde1f53ea06dc230519e9
+oid sha256:38f667f169ad688e25ac7c5407de2003cac6d1e4168f69a15fb61ba1a06681d7
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30baefd8be1f50d37d3bc31310171701a7ad8188fdc9a6584558407320f436dc
+oid sha256:0aac9303b6313dabec05560596d067d31e37caf3a02d16fdba574a9b8160764a
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:591a9c5e45bb1139399d3039b4c05dd34997d5462b11eb1ca1771e8044f4a6a1
+oid sha256:ab823abb115c32fa0673f758e0b0199bf6d5f3e43f5557dbadcc612729790b53
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1351871b6e069a7a4437616d40b913152ecc2cd046e996fcfeb300a3cf60638
+oid sha256:d397c26ac1ddd7670b9ddd8b909580a771b707ad5b586b657ad627e8bc4e787f
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.5539450645446777,
-  "best_model_checkpoint": "./outputs/checkpoint-1400",
-  "epoch": 1.0200364298724955,
+  "best_metric": 1.5348094701766968,
+  "best_model_checkpoint": "./outputs/checkpoint-1500",
+  "epoch": 1.092896174863388,
   "eval_steps": 100,
-  "global_step": 1400,
+  "global_step": 1500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -203,13 +203,27 @@
       "eval_samples_per_second": 21.794,
       "eval_steps_per_second": 2.727,
       "step": 1400
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0002,
+      "loss": 1.5249,
+      "step": 1500
+    },
+    {
+      "epoch": 1.09,
+      "eval_loss": 1.5348094701766968,
+      "eval_runtime": 287.891,
+      "eval_samples_per_second": 21.793,
+      "eval_steps_per_second": 2.727,
+      "step": 1500
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 8.347109656129536e+16,
+  "total_flos": 8.943146869488845e+16,
   "trial_name": null,
   "trial_params": null
 }