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:f5e93358dc84596ff9561b56bf06a60a527a5837b39695b4ab7d79634292e7bd
+oid sha256:1bf6e29139b0f24b399e976d09d4884b6235ffdaaa52ee8a0b7938baae37b428
 size 31466288

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9af4ca575718fe59e59e269281e9cb275bd99748b40f5662e0dc1c6081910eee
+oid sha256:a7a42b3115d3199d3743de0e8e6b98305c990ec30c405bfdddcdad84bc32c6e1
 size 62950917

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3893eedba13d04da0fc8309e22f0b6076e67fa50973b36efa761cf829a5f79f
+oid sha256:a6367fbbdff331ec434b7f226419eafab1e9622cbf8e5630817798ff44dfdb8a
 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": 1.1546565294265747,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.841530054644809,
+  "best_metric": 1.1375267505645752,
+  "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": 11.278,
       "eval_steps_per_second": 1.411,
       "step": 3900
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0002,
+      "loss": 1.0863,
+      "step": 4000
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 1.1375267505645752,
+      "eval_runtime": 556.442,
+      "eval_samples_per_second": 11.275,
+      "eval_steps_per_second": 1.411,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.0153734285909688e+18,
+  "total_flos": 1.0412060628345815e+18,
   "trial_name": null,
   "trial_params": null
 }