Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cdfba2964ce961719ae533a5b3885e6b255d0bf36c19035654780c1e1b5997c9
+oid sha256:708b87e047f76e1b90ceb5b1367b57f5282a029bbc30771c9df9444857706933
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ead8728468007a0027bcdab04f1d09712fe6a49caa3d4e7641bc47ca53ca7e32
+oid sha256:3563596ba1edce3e13c45ec1b125d5fa3788f9f25cc8f472c8f323ee2a3f1514
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19f566442a994eb367f21419c663f9d4a1cd6caaa8e2461d98745e40c3e8edf1
+oid sha256:2f8f8f48e0f8845915511c966c3679cf0755fa470bbf416453e88ad9a89e363e
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a42424f90eb68391066979b79f87ab18e2fce2a0abc640b46cf1b3f4c12c720a
+oid sha256:593f43b0c83adb27a2db37a6418c2ef12a213bbc2a02f2dc881de6846a69a931
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0015170574188232,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.841530054644809,
+  "best_metric": 1.9950693845748901,
+  "best_model_checkpoint": "./outputs/checkpoint-4000",
+  "epoch": 2.9143897996357016,
   "eval_steps": 100,
-  "global_step": 3900,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -553,13 +553,27 @@
       "eval_samples_per_second": 30.222,
       "eval_steps_per_second": 3.781,
       "step": 3900
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0002,
+      "loss": 1.9563,
+      "step": 4000
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 1.9950693845748901,
+      "eval_runtime": 208.0834,
+      "eval_samples_per_second": 30.151,
+      "eval_steps_per_second": 3.773,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1378537348161536e+17,
+  "total_flos": 1.1667957018181632e+17,
   "trial_name": null,
   "trial_params": null
 }