Training in progress, step 900, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:867e03a4fd3f23e4f6a08d96561e2f5f283004214293470951129ebf31e3c054
+oid sha256:4269d3b9b509e5eed5409c2e16fd7d195914f88586d1ad09ccc74ded67b204a5
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc91af28b835c85175bd7530fef83335620c114d04ec7194c95e14ec73146e02
+oid sha256:d0f6b2f0ccfa41cd106da53f5e14bf7d59953ae5d9323f2a9877b8417e5f496b
 size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abaef124d8e265d68d5751114b7c7b99ba0eb280d27af45a9884fb7e66277ae9
+oid sha256:99f1c1111b8d1f0cb2cc4b559a49a0a42b0bb5df350eda9361510f1ff2ab08e2
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de7840bcb72f2f480fd301578d289cdfa174589e831b0d33e5772f3956b6beae
+oid sha256:a3c39113390004215769b6e7fbe73f1073dd2faa3ab8f9ff815b62e9c2f80c82
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.6823335886001587,
-  "best_model_checkpoint": "./outputs/checkpoint-800",
-  "epoch": 0.5828779599271403,
+  "best_metric": 1.6567240953445435,
+  "best_model_checkpoint": "./outputs/checkpoint-900",
+  "epoch": 0.6557377049180327,
   "eval_steps": 100,
-  "global_step": 800,
+  "global_step": 900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -119,13 +119,27 @@
       "eval_samples_per_second": 43.565,
       "eval_steps_per_second": 5.451,
       "step": 800
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0002,
+      "loss": 1.6619,
+      "step": 900
+    },
+    {
+      "epoch": 0.66,
+      "eval_loss": 1.6567240953445435,
+      "eval_runtime": 144.0655,
+      "eval_samples_per_second": 43.55,
+      "eval_steps_per_second": 5.449,
+      "step": 900
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 4.770964776085094e+16,
+  "total_flos": 5.365902922462003e+16,
   "trial_name": null,
   "trial_params": null
 }