Training in progress, step 3700, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fcd8ce5098ca17c84c555c34a24ddbf246492c4f81a98f86837e33372705e9da
+oid sha256:78e4a843dcfb85dc56eab8279740c9272f8ad8e83fe9c9291499b1ca255e915a
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75321fc617a4f566758408da632c978bfba3277fb3068233e8744e870ec37b61
+oid sha256:3a6892ae9c8bf7f994967575c5a1109ea4dd2332e6ffa8607e773ebc53535c0d
 size 18902665

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfe63e551ea8e601e6b37db6f5c29b9e1b4d893ef2a031f35638bcd71863bbde
+oid sha256:d24dae49552cd627e5eaaed4e33ff7d215a6f396899031132e935859a4b208cd
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43e3cf8d56a3f083d00cc85544d76ada2f884a1018c8752332d96f2799911117
+oid sha256:c157d4508adfa804f177e78fb4986d5f3ffb3ec20ae255fb757aa13c0595fdb4
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.0046775341033936,
-  "best_model_checkpoint": "./outputs/checkpoint-3600",
-  "epoch": 2.6229508196721314,
+  "best_metric": 1.9959838390350342,
+  "best_model_checkpoint": "./outputs/checkpoint-3700",
+  "epoch": 2.6958105646630237,
   "eval_steps": 100,
-  "global_step": 3600,
+  "global_step": 3700,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -511,13 +511,27 @@
       "eval_samples_per_second": 30.436,
       "eval_steps_per_second": 3.808,
       "step": 3600
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.0002,
+      "loss": 1.9565,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 1.9959838390350342,
+      "eval_runtime": 214.3631,
+      "eval_samples_per_second": 29.268,
+      "eval_steps_per_second": 3.662,
+      "step": 3700
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.057199681212416e+17,
+  "total_flos": 1.0863248925892608e+17,
   "trial_name": null,
   "trial_params": null
 }