Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72a718909f6cbcec9e9779ea7c764692106bd15e80003010a1b55f39f8f73f90
+oid sha256:961ea186a47b7ade49109754c046bc573efd9aaafb86ab434362c8f776a9a5e1
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3aa8aa20fcc9d2fc1b8425448da917b1a9c8e220737bbc024cef55c1e498f9dd
+oid sha256:10a16a6a07f3ed33f0d12bf76835917926a369d20c2fed330a119dfbdf65a339
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c70a24ac9111081d6a82fd74b326ecd79952e484fad273fd13a8f86fecde382
+oid sha256:7dabce81359fa612462d377e39b2983dee4746040809580c50e847ae360e461b
 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": 1.2307173013687134,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.841894353369763,
+  "best_metric": 1.2222630977630615,
+  "best_model_checkpoint": "./outputs/checkpoint-4000",
+  "epoch": 2.914754098360656,
   "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": 46.886,
       "eval_steps_per_second": 5.866,
       "step": 3900
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0002,
+      "loss": 1.2025,
+      "step": 4000
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 1.2222630977630615,
+      "eval_runtime": 133.8615,
+      "eval_samples_per_second": 46.869,
+      "eval_steps_per_second": 5.864,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.324643974858588e+17,
+  "total_flos": 2.3841325139747635e+17,
   "trial_name": null,
   "trial_params": null
 }