Training in progress, step 3600, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7b04e7516d0b91064c34bdcf2a76c71b61f453d13cc129450d1c2bfe11747dbd
+oid sha256:18b1b00377fb4beb8e5f4a8606838ad99fb8f485aacf41ade174a9109fa5e721
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab2ffb23a2335b8790c2fd1144818605bc24c50fba5802fbbe6357d08235b0ad
+oid sha256:7007467fd329997eac8d988aa2ad87c66b062240a71cd2519e3f3d51253bae72
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5be7640a4620dca9ae692e481495dbb2e47e1c1397961515028b5584ebe5eb54
+oid sha256:070f30b5eb20ec8e39954e94771174ba38f14b02759b5ed410e96e86ca5a144e
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31f466e63f2d702a9646f61d3cb0499d7a443ca833cfea51694a53eaa24cfd01
+oid sha256:43e3cf8d56a3f083d00cc85544d76ada2f884a1018c8752332d96f2799911117
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0399410724639893,
-  "best_model_checkpoint": "./outputs/checkpoint-3500",
-  "epoch": 2.5500910746812386,
+  "best_metric": 2.033215284347534,
+  "best_model_checkpoint": "./outputs/checkpoint-3600",
+  "epoch": 2.6229508196721314,
   "eval_steps": 100,
-  "global_step": 3500,
+  "global_step": 3600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -497,13 +497,27 @@
       "eval_samples_per_second": 30.267,
       "eval_steps_per_second": 3.787,
       "step": 3500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0002,
+      "loss": 1.9977,
+      "step": 3600
+    },
+    {
+      "epoch": 2.62,
+      "eval_loss": 2.033215284347534,
+      "eval_runtime": 207.459,
+      "eval_samples_per_second": 30.242,
+      "eval_steps_per_second": 3.784,
+      "step": 3600
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.02120256014336e+17,
+  "total_flos": 1.0503452935471104e+17,
   "trial_name": null,
   "trial_params": null
 }