Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c66c167440aeb0946d6279663e50d9312ceb473cdcf6f29f4160f36a159995c
+oid sha256:b2c7e59797ad72d80ee2c9d28ef09971742ce54beb467a8d5c5a022a4f6ea4c5
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e556816dc43ae052927f85f00760ef86b74653fc28d4c08a26ef67b4c14e5be
+oid sha256:23234c009b455307cd85906aa6108ec039edea250585556b5f0dbbc8e683f7d3
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f387f966aecfa76017d726e43b1fa9c3ccecd8af5742b255e2ef66c7eeb645b
+oid sha256:a9ef6c3858e7d84aa5f4da6d612efb7619c9f1a991da1e501295c2a72877c544
 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": 2.069366693496704,
-  "best_model_checkpoint": "./outputs/checkpoint-3900",
-  "epoch": 2.841530054644809,
+  "best_metric": 2.0633840560913086,
+  "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": 30.547,
       "eval_steps_per_second": 3.822,
       "step": 3900
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0002,
+      "loss": 2.0305,
+      "step": 4000
+    },
+    {
+      "epoch": 2.91,
+      "eval_loss": 2.0633840560913086,
+      "eval_runtime": 205.8541,
+      "eval_samples_per_second": 30.478,
+      "eval_steps_per_second": 3.813,
+      "step": 4000
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1105217882630144e+17,
+  "total_flos": 1.1387458434336768e+17,
   "trial_name": null,
   "trial_params": null
 }