Training in progress, step 2100, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e9e419fbca9a934034d0a82cf57b8ef7f012cff92f23371e54a9b1cb99cd101
+oid sha256:5cd9d0525f4443a3c492bdda33a70669036edfcadab1ed9c3f6203802eba3bd9
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9604cc61a44f05f3ed132da7d6a787bae655101094ae779a83ef65a6a0d2222
+oid sha256:2d59cfc8c9feb07bdb9d01b40604d8f3ea34f48224b012f12867179695083de0
 size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c134cfa86adc7fd0d893de0ed5f1d042c69cb03ed5bb956e86d5d878b172fe1
+oid sha256:347cad9f3bce277eed003720972c437ba21dbbd4b008635e9d49d0ca58e0dcb4
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:720bafe9a34b3041027dccef7d9c03bb23064e1fcdd80ae5f27e717d41b7cf31
+oid sha256:b1238b3bfa75e49a19396161e9e7b72ab89cdd1a3f63b51c0ab4d6e8d216c5a5
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.578723669052124,
-  "best_model_checkpoint": "./outputs/checkpoint-2000",
-  "epoch": 1.4571948998178508,
+  "best_metric": 1.562243938446045,
+  "best_model_checkpoint": "./outputs/checkpoint-2100",
+  "epoch": 1.530054644808743,
   "eval_steps": 100,
-  "global_step": 2000,
+  "global_step": 2100,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -287,13 +287,27 @@
       "eval_samples_per_second": 43.549,
       "eval_steps_per_second": 5.449,
       "step": 2000
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 0.0002,
+      "loss": 1.5572,
+      "step": 2100
+    },
+    {
+      "epoch": 1.53,
+      "eval_loss": 1.562243938446045,
+      "eval_runtime": 144.0023,
+      "eval_samples_per_second": 43.569,
+      "eval_steps_per_second": 5.451,
+      "step": 2100
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1910623325700915e+17,
+  "total_flos": 1.2505986444644352e+17,
   "trial_name": null,
   "trial_params": null
 }