Training in progress, step 3800, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:506c9daa9a63f8a0359134f35e7222923af01f5861d09d9b702197ca38eff035
+oid sha256:38acc6e27c9e910592a3928094223ff54a296092f8bcb05d0ad3a8b0184c65c9
 size 50338848

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ebe1a670b9bd5b8db81a448d712a1caa862846faee9fbbfbb84cdc51054a1a9b
+oid sha256:8fb0ecdec215ce19ff9fa28f7e8a359fc57fec9788315b4eb61ac30cac2280e8
 size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff244788f228d59f20add1a127912b5b144f70e4c321b25031a681809deb1ae0
+oid sha256:0625d44593f9a35a65bd6a0c353b2ee783deaa835a908009e34809ad32c4166a
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c157d4508adfa804f177e78fb4986d5f3ffb3ec20ae255fb757aa13c0595fdb4
+oid sha256:3961d2f201ce99426ee113655da78d8214f6ef72fc082855daa3c565989888c2
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.255784034729004,
-  "best_model_checkpoint": "./outputs/checkpoint-3700",
-  "epoch": 2.6958105646630237,
+  "best_metric": 1.2465852499008179,
+  "best_model_checkpoint": "./outputs/checkpoint-3800",
+  "epoch": 2.768670309653916,
   "eval_steps": 100,
-  "global_step": 3700,
+  "global_step": 3800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -525,13 +525,27 @@
       "eval_samples_per_second": 43.556,
       "eval_steps_per_second": 5.45,
       "step": 3700
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0002,
+      "loss": 1.2112,
+      "step": 3800
+    },
+    {
+      "epoch": 2.77,
+      "eval_loss": 1.2465852499008179,
+      "eval_runtime": 144.0134,
+      "eval_samples_per_second": 43.565,
+      "eval_steps_per_second": 5.451,
+      "step": 3800
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.2052180376706253e+17,
+  "total_flos": 2.2648457919379046e+17,
   "trial_name": null,
   "trial_params": null
 }