Training in progress, step 500, checkpoint

last-checkpoint/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:943eae222e4add77c1ddafafb0d45afca7eb01031f9a77fe679c78ee379d7177
+oid sha256:433015265cd1b3ee919b046133931fbbf412c76b2126bba58655a0b670c00a14
 size 50349441

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ef9cc71e325a5462fe060cbbdf5546a243c527c38f6436dc0d8634b34c83e2df
+oid sha256:bba530906d55e08e55a6b78b5ba3bb275638c8737f743616f1e1865e13573642
 size 100693001

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b17127c1713807334f935edb094b8a47aee68e40641e90a7299725b5a6682dd6
+oid sha256:52d2d6e14904128a443b518d8e76a7926f8a1af0669bd545e4ac0538a858cbbe
 size 14575

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13276f15dd2b6acc19b970176aa2db4ac9b58241843e72c89b50e3094e903b19
+oid sha256:5efc416a6883409dd7ab6f5c779e107c7c2baa7af6e12ed9fbd9dd73b8b20784
 size 627

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.8125503063201904,
-  "best_model_checkpoint": "./outputs/checkpoint-400",
-  "epoch": 0.29143897996357016,
+  "best_metric": 1.7720445394515991,
+  "best_model_checkpoint": "./outputs/checkpoint-500",
+  "epoch": 0.36429872495446264,
   "eval_steps": 100,
-  "global_step": 400,
+  "global_step": 500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -63,13 +63,27 @@
       "eval_samples_per_second": 43.574,
       "eval_steps_per_second": 5.452,
       "step": 400
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.0002,
+      "loss": 1.785,
+      "step": 500
+    },
+    {
+      "epoch": 0.36,
+      "eval_loss": 1.7720445394515991,
+      "eval_runtime": 144.0061,
+      "eval_samples_per_second": 43.568,
+      "eval_steps_per_second": 5.451,
+      "step": 500
     }
   ],
   "logging_steps": 100,
   "max_steps": 4116,
   "num_train_epochs": 3,
   "save_steps": 100,
-  "total_flos": 2.386997635055616e+16,
+  "total_flos": 2.9834861985890304e+16,
   "trial_name": null,
   "trial_params": null
 }