Training in progress, step 300, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:746165a2764849c8c6dbde4fda8fe622a44314bd48f356b3e382f5155ed976b3
+oid sha256:efa49ae50a77b7d26a8c97f9fe172cc5909cd105dcbccc6177275c6d90989768
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d97d38cb88545389354c4173ee03c29c858dfb4d72c7c126d90e1d145c165349
-size 100692937
+oid sha256:10b49364065c45fc89500c5d0792917a4c380cb14f3ef0d33b6c05f698b96802
+size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8cca4050b412069a5a574ef82deadd3fd628d6acb7444705fb5ae44f31ae444
+oid sha256:0991493e3ac522258a52c5e5855e34a1077a108bf34c050d61cd9082a802fb6b
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7aa2c8b84e17817e6a4dcba5955fca913e266fdcd47f5594a29933ebd4972a01
+oid sha256:c3e02e4b8a74e841ee5a1ce8c453a9becb15f5be6cbcdd100b814e93235dab36
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.9195271730422974,
-  "best_model_checkpoint": "./outputs/checkpoint-200",
-  "epoch": 0.14571948998178508,
+  "best_metric": 1.858821988105774,
+  "best_model_checkpoint": "./outputs/checkpoint-300",
+  "epoch": 0.2185792349726776,
   "eval_steps": 100,
-  "global_step": 200,
+  "global_step": 300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -35,13 +35,27 @@
       "eval_samples_per_second": 43.501,
       "eval_steps_per_second": 5.443,
       "step": 200
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002,
+      "loss": 1.8881,
+      "step": 300
+    },
+    {
+      "epoch": 0.22,
+      "eval_loss": 1.858821988105774,
+      "eval_runtime": 144.0313,
+      "eval_samples_per_second": 43.56,
+      "eval_steps_per_second": 5.45,
+      "step": 300
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 1.1917344486653952e+16,
+  "total_flos": 1.7888590056259584e+16,
   "trial_name": null,
   "trial_params": null
 }