Training in progress, step 3700, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7bd11fbc65a2130792b8cf788b9f9ddd061adb7bd3d546548d5a2e1ed21360a5
+oid sha256:09d4b91e1e9659278f7bba665d6e7114c775bcc0952d8abdb4b02198254f6920
 size 9444296

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:628db2b0870fe25de936d9c9683b1cf172d9d3ed7862d689c5ac355cd2b5d3bf
+oid sha256:f5140c747c623d47fd8a47dd5498f26aec184782d67d4562e667b045ba4f2c19
 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": 2.0044028759002686,
-  "best_model_checkpoint": "./outputs/checkpoint-3600",
-  "epoch": 2.6229508196721314,
+  "best_metric": 1.994431972503662,
+  "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": 30.019,
       "eval_steps_per_second": 3.756,
       "step": 3600
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.0002,
+      "loss": 1.9572,
+      "step": 3700
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 1.994431972503662,
+      "eval_runtime": 209.7442,
+      "eval_samples_per_second": 29.913,
+      "eval_steps_per_second": 3.743,
+      "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
 }