Training in progress, step 3600, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7816e8f7fe1531fa69378cef19ec460f7cb146ae3ae4dc1099c995c8e852db03
+oid sha256:5b47f7cb7f80077d04d264b8f37193d2acf6798c4a211d0ffff9d7ea8e53d8c0
 size 75507072

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c74a5d8dd123d02ca70d0431b4750fba5d3d6397e15c9ba4d61293adc46352d1
+oid sha256:3f9ba58e13cbad9ee95bcbf4071bc0eed8ec6de53092d6b862bbe300f43bd3b3
 size 151032837

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0df161def7cd001c4c87cef77ffd46a655436a9551b23f43fa5264ada38a0aff
+oid sha256:baca4f1b68fe3db55c3f9c8ea267473af8c9b358050255190a124d44cbf1200b
 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": 1.1449061632156372,
-  "best_model_checkpoint": "./outputs/checkpoint-3500",
-  "epoch": 2.550819672131148,
+  "best_metric": 1.1353813409805298,
+  "best_model_checkpoint": "./outputs/checkpoint-3600",
+  "epoch": 2.62367941712204,
   "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": 18.472,
       "eval_steps_per_second": 2.311,
       "step": 3500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0002,
+      "loss": 1.116,
+      "step": 3600
+    },
+    {
+      "epoch": 2.62,
+      "eval_loss": 1.1353813409805298,
+      "eval_runtime": 340.0732,
+      "eval_samples_per_second": 18.449,
+      "eval_steps_per_second": 2.308,
+      "step": 3600
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1348301669487411e+18,
+  "total_flos": 1.167321484023469e+18,
   "trial_name": null,
   "trial_params": null
 }