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:b5bd53816dec1f54087c7a98a07ce3e939ca34b4f4b6d19c7d7deda012989e55
+oid sha256:be87eb6315475e805ac0626f8bc07530bee80f088c7382143adee24ae01c4cdb
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00177b501de8ddd7c362d6cf0908f8399182383eaec609ad095a19b45401d4a1
+oid sha256:a9b4497f0e0bf6293beae82e3e73cca6eb1509573193b0eccb1aafdc56f8d449
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:34985e8b1e69a205583c6b9115293e324729ed0c191fbcce181660393717f45d
+oid sha256:586ee95ec4b1c09d119f8cf73ef46748706d8ebb242b12344dfbad2ef9da6ebf
 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": 1.340726613998413,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.8422586520947175,
+  "best_metric": 1.3258968591690063,
+  "best_model_checkpoint": "./outputs/checkpoint-4000",
+  "epoch": 2.9151183970856103,
   "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": 47.176,
       "eval_steps_per_second": 5.903,
       "step": 3900
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.0002,
+      "loss": 1.2963,
+      "step": 4000
+    },
+    {
+      "epoch": 2.92,
+      "eval_loss": 1.3258968591690063,
+      "eval_runtime": 132.9668,
+      "eval_samples_per_second": 47.185,
+      "eval_steps_per_second": 5.904,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.3221217626762445e+17,
+  "total_flos": 2.3816651085992755e+17,
   "trial_name": null,
   "trial_params": null
 }