Training in progress, step 4100, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c138f9e07b2bd6b8cc28592584319298f4af0599c517b236106d59cf40af67e
+oid sha256:3006906b1291956b66d034d42d8db67ce209517f97b2b8aedca2bdcc6ef48f92
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:adebca319d4bef557a07ca8008697da3a127dd09d3404287ab6aeb0d30f777fa
+oid sha256:40aff8f8a11b475717989a53327db264a4c6be1dce8be2027dec6a61bec43b9e
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe072b43116ea6af1d9839ff9078df37c5fb13f3285b4cbe739c3c5d14228bdf
+oid sha256:8a0d9d4b9143f23093465c736b344edb94f3988c049bd9c6571469763fcf7a30
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:593f43b0c83adb27a2db37a6418c2ef12a213bbc2a02f2dc881de6846a69a931
+oid sha256:f55ebe991e4838deda9ff3044672335e03697b65b3fe0753cca61998e974ee51
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.9696241617202759,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9143897996357016,
+  "best_metric": 1.9596437215805054,
+  "best_model_checkpoint": "./outputs/checkpoint-4100",
+  "epoch": 2.987249544626594,
   "eval_steps": 100,
-  "global_step": 4000,
+  "global_step": 4100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -567,13 +567,27 @@
       "eval_samples_per_second": 14.201,
       "eval_steps_per_second": 1.777,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 1.9403,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.9596437215805054,
+      "eval_runtime": 353.6394,
+      "eval_samples_per_second": 17.741,
+      "eval_steps_per_second": 2.22,
+      "step": 4100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1744177835220992e+17,
+  "total_flos": 1.2037527679500288e+17,
   "trial_name": null,
   "trial_params": null
 }