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:be87eb6315475e805ac0626f8bc07530bee80f088c7382143adee24ae01c4cdb
+oid sha256:39c43ba531dd049bf9085b4d6e4270d0b9c6a927c2726c5081e06c6f6e4e97ec
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9b4497f0e0bf6293beae82e3e73cca6eb1509573193b0eccb1aafdc56f8d449
+oid sha256:8f10f68e3e86c2b511eeec0821249bde457f69e65f64a9af68431e3a1ae15ae1
 size 100691721

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:586ee95ec4b1c09d119f8cf73ef46748706d8ebb242b12344dfbad2ef9da6ebf
+oid sha256:2761db6e438999d2c6b14fa0a45a58a86ef8f9b94453fe6c5383e648441625d5
 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": 1.3258968591690063,
-  "best_model_checkpoint": "./outputs/checkpoint-4000",
-  "epoch": 2.9151183970856103,
+  "best_metric": 1.3161581754684448,
+  "best_model_checkpoint": "./outputs/checkpoint-4100",
+  "epoch": 2.9879781420765026,
   "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": 47.185,
       "eval_steps_per_second": 5.904,
       "step": 4000
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 0.0002,
+      "loss": 1.2867,
+      "step": 4100
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.3161581754684448,
+      "eval_runtime": 132.905,
+      "eval_samples_per_second": 47.207,
+      "eval_steps_per_second": 5.906,
+      "step": 4100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.3816651085992755e+17,
+  "total_flos": 2.4411263908542874e+17,
   "trial_name": null,
   "trial_params": null
 }