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:dde893a5b8c9903703f5fe18b4db6a74d80d1e605a1654c710070dd9d9ede78f
+oid sha256:6f654fa6534ae48416fd3b50f90a80eafdeed68b33a32a6aace4caab3df83a1c
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df8a418b0d49534a4c1e36fe0283b8c460e6713524b33bd3cf5f0b2bae83bf5e
+oid sha256:0e3ac252efb905580c340792dde23762e62e7c716fc2902b717985412f24587c
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15ffa04f9b4c47e23e16441fa376bd921a73b836cf43b7e6e2b1582f10fd2d81
+oid sha256:5840f7374f51c709f33701383022a2fde86c49fec768033add729bc923ee95d3
 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": 1.2287747859954834,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.841894353369763,
+  "best_metric": 1.2198114395141602,
+  "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": 47.006,
       "eval_steps_per_second": 5.881,
       "step": 3900
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0002,
+      "loss": 1.1995,
+      "step": 4000
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 1.2198114395141602,
+      "eval_runtime": 133.4977,
+      "eval_samples_per_second": 46.997,
+      "eval_steps_per_second": 5.88,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.3246378200834867e+17,
+  "total_flos": 2.384126359199662e+17,
   "trial_name": null,
   "trial_params": null
 }