Training in progress, epoch 0

all_results.json

@@ -7,7 +7,7 @@
     "eval_steps_per_second": 1.417,
     "total_flos": 2.039607446848635e+17,
     "train_loss": 0.0,
-    "train_runtime": 162.0706,
-    "train_samples_per_second": 16.659,
-    "train_steps_per_second": 0.13
+    "train_runtime": 134.7541,
+    "train_samples_per_second": 20.037,
+    "train_steps_per_second": 0.156
 }

runs/Jul27_11-16-03_aaa288b39908/events.out.tfevents.1690456571.aaa288b39908.1427.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca2202099ecaf4a716c63a684e55a569a0d98fc4311260ce8964f532b87f3fae
+size 4752

train_results.json

@@ -2,7 +2,7 @@
     "epoch": 2.9,
     "total_flos": 2.039607446848635e+17,
     "train_loss": 0.0,
-    "train_runtime": 162.0706,
-    "train_samples_per_second": 16.659,
-    "train_steps_per_second": 0.13
+    "train_runtime": 134.7541,
+    "train_samples_per_second": 20.037,
+    "train_steps_per_second": 0.156
 }

trainer_state.json

@@ -11,9 +11,9 @@
       "epoch": 0.97,
       "eval_accuracy": 0.12,
       "eval_loss": NaN,
-      "eval_runtime": 5.9456,
-      "eval_samples_per_second": 37.843,
-      "eval_steps_per_second": 1.346,
+      "eval_runtime": 5.4828,
+      "eval_samples_per_second": 41.038,
+      "eval_steps_per_second": 1.459,
       "step": 7
     },
     {
@@ -26,9 +26,9 @@
       "epoch": 1.93,
       "eval_accuracy": 0.12,
       "eval_loss": NaN,
-      "eval_runtime": 5.5037,
-      "eval_samples_per_second": 40.882,
-      "eval_steps_per_second": 1.454,
+      "eval_runtime": 6.3386,
+      "eval_samples_per_second": 35.497,
+      "eval_steps_per_second": 1.262,
       "step": 14
     },
     {
@@ -41,9 +41,9 @@
       "epoch": 2.9,
       "eval_accuracy": 0.12,
       "eval_loss": NaN,
-      "eval_runtime": 6.3715,
-      "eval_samples_per_second": 35.313,
-      "eval_steps_per_second": 1.256,
+      "eval_runtime": 6.7326,
+      "eval_samples_per_second": 33.419,
+      "eval_steps_per_second": 1.188,
       "step": 21
     },
     {
@@ -51,9 +51,9 @@
       "step": 21,
       "total_flos": 2.039607446848635e+17,
       "train_loss": 0.0,
-      "train_runtime": 162.0706,
-      "train_samples_per_second": 16.659,
-      "train_steps_per_second": 0.13
+      "train_runtime": 134.7541,
+      "train_samples_per_second": 20.037,
+      "train_steps_per_second": 0.156
     }
   ],
   "max_steps": 21,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92c41bebb28e088071a32a0cbf4a2aec358acfac5b21a7692231bb3122cafb93
-size 4027
+oid sha256:be96c993823229264acb900936ec766db5b0aae7150ff38bfc6ecea4b93fc655
+size 4155