Model save

all_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 0.0,
     "train_loss": 0.6931471824645996,
-    "train_runtime": 22.2221,
-    "train_samples_per_second": 2.88,
-    "train_steps_per_second": 0.045
+    "train_runtime": 5.3639,
+    "train_samples_per_second": 11.932,
+    "train_steps_per_second": 0.186
 }

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 0.0,
     "train_loss": 0.6931471824645996,
-    "train_runtime": 22.2221,
-    "train_samples_per_second": 2.88,
-    "train_steps_per_second": 0.045
+    "train_runtime": 5.3639,
+    "train_samples_per_second": 11.932,
+    "train_steps_per_second": 0.186
 }

trainer_state.json

@@ -11,10 +11,10 @@
     {
       "epoch": 0.0,
       "learning_rate": 0.0,
-      "logits/generated": -1.3129560947418213,
-      "logits/real": -0.6997354626655579,
-      "logps/generated": -609.5880126953125,
-      "logps/real": -542.52783203125,
+      "logits/generated": -1.3131200075149536,
+      "logits/real": -0.6672423481941223,
+      "logps/generated": -604.0491943359375,
+      "logps/real": -541.8375854492188,
       "loss": 0.6931,
       "rewards/accuracies": 0.0,
       "rewards/generated": 0.0,
@@ -27,9 +27,9 @@
       "step": 1,
       "total_flos": 0.0,
       "train_loss": 0.6931471824645996,
-      "train_runtime": 22.2221,
-      "train_samples_per_second": 2.88,
-      "train_steps_per_second": 0.045
+      "train_runtime": 5.3639,
+      "train_samples_per_second": 11.932,
+      "train_steps_per_second": 0.186
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b76d577a3000f391b9d71fe4f8351365647d08ebe72ece646311115d02467530
-size 5944
+oid sha256:6d8b6e54292bea39dbb8cbd18ed891032caf6bd8faede14faf65a67dbf8a6418
+size 4792