Update model files

all_results.json

@@ -1,15 +1,15 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.5337530906790869,
-    "eval_loss": 2.223806858062744,
-    "eval_runtime": 2.3003,
-    "eval_samples": 34,
-    "eval_samples_per_second": 14.781,
-    "eval_steps_per_second": 3.043,
-    "perplexity": 9.242448669461155,
-    "train_loss": 2.4565388997395834,
-    "train_runtime": 673.0137,
-    "train_samples": 178,
-    "train_samples_per_second": 5.29,
-    "train_steps_per_second": 1.07
+    "eval_accuracy": 0.5646085968666614,
+    "eval_loss": 2.086466073989868,
+    "eval_runtime": 2.4776,
+    "eval_samples": 37,
+    "eval_samples_per_second": 14.934,
+    "eval_steps_per_second": 3.229,
+    "perplexity": 8.056394097668248,
+    "train_loss": 2.272189174554287,
+    "train_runtime": 742.3325,
+    "train_samples": 193,
+    "train_samples_per_second": 5.2,
+    "train_steps_per_second": 1.051
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.5337530906790869,
-    "eval_loss": 2.223806858062744,
-    "eval_runtime": 2.3003,
-    "eval_samples": 34,
-    "eval_samples_per_second": 14.781,
-    "eval_steps_per_second": 3.043,
-    "perplexity": 9.242448669461155
+    "eval_accuracy": 0.5646085968666614,
+    "eval_loss": 2.086466073989868,
+    "eval_runtime": 2.4776,
+    "eval_samples": 37,
+    "eval_samples_per_second": 14.934,
+    "eval_steps_per_second": 3.229,
+    "perplexity": 8.056394097668248
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:28ef587b6f2cf44fcf4ae525a4d4a1844c46ead2716ff662e84d68e944850160
+oid sha256:3cf454e9c11d7843feb05895cd1bf4b9eefd03a5c7584788d4fec6ee7624af81
 size 510398013

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
-    "train_loss": 2.4565388997395834,
-    "train_runtime": 673.0137,
-    "train_samples": 178,
-    "train_samples_per_second": 5.29,
-    "train_steps_per_second": 1.07
+    "train_loss": 2.272189174554287,
+    "train_runtime": 742.3325,
+    "train_samples": 193,
+    "train_samples_per_second": 5.2,
+    "train_steps_per_second": 1.051
 }

trainer_state.json

@@ -2,30 +2,30 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 20.0,
-  "global_step": 720,
+  "global_step": 780,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 13.89,
-      "learning_rate": 1.527777777777778e-05,
-      "loss": 2.6405,
+      "epoch": 12.82,
+      "learning_rate": 1.794871794871795e-05,
+      "loss": 2.4721,
       "step": 500
     },
     {
       "epoch": 20.0,
-      "step": 720,
-      "total_flos": 1860399267840000.0,
-      "train_loss": 2.4565388997395834,
-      "train_runtime": 673.0137,
-      "train_samples_per_second": 5.29,
-      "train_steps_per_second": 1.07
+      "step": 780,
+      "total_flos": 2017174487040000.0,
+      "train_loss": 2.272189174554287,
+      "train_runtime": 742.3325,
+      "train_samples_per_second": 5.2,
+      "train_steps_per_second": 1.051
     }
   ],
-  "max_steps": 720,
+  "max_steps": 780,
   "num_train_epochs": 20,
-  "total_flos": 1860399267840000.0,
+  "total_flos": 2017174487040000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00c461014766e6880f76012958d9e671b69b5f4ec251d415d9a43f10fc8770ac
+oid sha256:ed95b98baa6effbfb20f5899a83a69635492a9d86ff1bbb1ffc8b9821eed8097
 size 3515