Training in progress, epoch 1

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.6787003610108303,
-    "eval_loss": 0.610024094581604,
-    "eval_runtime": 0.2213,
+    "eval_accuracy": 0.7220216606498195,
+    "eval_loss": 0.5765948295593262,
+    "eval_runtime": 0.2028,
     "eval_samples": 277,
-    "eval_samples_per_second": 1251.949,
-    "eval_steps_per_second": 13.559,
-    "train_loss": 0.5981983282627203,
-    "train_runtime": 104.8491,
+    "eval_samples_per_second": 1365.807,
+    "eval_steps_per_second": 14.792,
+    "train_loss": 0.5880965404021434,
+    "train_runtime": 74.518,
     "train_samples": 2490,
-    "train_samples_per_second": 237.484,
-    "train_steps_per_second": 7.439
+    "train_samples_per_second": 334.147,
+    "train_steps_per_second": 10.467
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.6787003610108303,
-    "eval_loss": 0.610024094581604,
-    "eval_runtime": 0.2213,
+    "eval_accuracy": 0.7220216606498195,
+    "eval_loss": 0.5765948295593262,
+    "eval_runtime": 0.2028,
     "eval_samples": 277,
-    "eval_samples_per_second": 1251.949,
-    "eval_steps_per_second": 13.559
+    "eval_samples_per_second": 1365.807,
+    "eval_steps_per_second": 14.792
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4db62de3019dbbb8c55885f95313dc4d45ec3333715d8510f7ad220b39cf982c
-size 2667673
+oid sha256:4a4db7b252735a0caa971bc9b8bfcc416a7e035a6058ce037760f7e1a58c5a81
+size 2704633

runs/Jan29_14-47-20_node4/events.out.tfevents.1643438860.node4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bac5fc32b72bfebbdbf921e14aebd6d0bf4931baaf404966f1d3ea06f2ac4611
-size 4479
+oid sha256:0c2fd211d0b1c2327beb4d58329becb516127a4f116e95f8c4ea26ba8eaed696
+size 7251

runs/Jan29_14-47-20_node4/events.out.tfevents.1643438935.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b31e175e0dd15cbfea3318f4bfb0e0a83405b6985f58e848789a2e33605a92f
+size 363

runs/Jan29_15-48-48_node4/1643442552.7518203/events.out.tfevents.1643442552.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:034979261d00c32eea5f23ba24c3987c8a6f24cbd130f79847c0673ce148d878
+size 4585

runs/Jan29_15-48-48_node4/events.out.tfevents.1643442552.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:152959d8f445f523b1ae99917d4d57fe578da9f5ae4f4ffe929d6ec220408882
+size 3515

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 0.5981983282627203,
-    "train_runtime": 104.8491,
+    "train_loss": 0.5880965404021434,
+    "train_runtime": 74.518,
     "train_samples": 2490,
-    "train_samples_per_second": 237.484,
-    "train_steps_per_second": 7.439
+    "train_samples_per_second": 334.147,
+    "train_steps_per_second": 10.467
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.6787003610108303,
-  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-546",
+  "best_metric": 0.7220216606498195,
+  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-780",
   "epoch": 10.0,
   "global_step": 780,
   "is_hyper_param_search": false,
@@ -10,107 +10,107 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.7017449736595154,
-      "eval_runtime": 0.2158,
-      "eval_samples_per_second": 1283.674,
-      "eval_steps_per_second": 13.903,
+      "eval_loss": 0.7042641639709473,
+      "eval_runtime": 0.2314,
+      "eval_samples_per_second": 1197.187,
+      "eval_steps_per_second": 12.966,
       "step": 78
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5054151624548736,
-      "eval_loss": 0.6921688318252563,
-      "eval_runtime": 0.196,
-      "eval_samples_per_second": 1412.998,
-      "eval_steps_per_second": 15.303,
+      "eval_accuracy": 0.5776173285198556,
+      "eval_loss": 0.6811862587928772,
+      "eval_runtime": 0.2009,
+      "eval_samples_per_second": 1378.756,
+      "eval_steps_per_second": 14.932,
       "step": 156
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.555956678700361,
-      "eval_loss": 0.66964191198349,
-      "eval_runtime": 0.2062,
-      "eval_samples_per_second": 1343.15,
-      "eval_steps_per_second": 14.547,
+      "eval_accuracy": 0.6028880866425993,
+      "eval_loss": 0.6594843864440918,
+      "eval_runtime": 0.2361,
+      "eval_samples_per_second": 1173.317,
+      "eval_steps_per_second": 12.707,
       "step": 234
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6353790613718412,
-      "eval_loss": 0.6405073404312134,
-      "eval_runtime": 0.1941,
-      "eval_samples_per_second": 1427.34,
-      "eval_steps_per_second": 15.459,
+      "eval_accuracy": 0.6714801444043321,
+      "eval_loss": 0.6216505765914917,
+      "eval_runtime": 0.1961,
+      "eval_samples_per_second": 1412.672,
+      "eval_steps_per_second": 15.3,
       "step": 312
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6209386281588448,
-      "eval_loss": 0.6625295877456665,
-      "eval_runtime": 0.195,
-      "eval_samples_per_second": 1420.282,
-      "eval_steps_per_second": 15.382,
+      "eval_accuracy": 0.6787003610108303,
+      "eval_loss": 0.6077037453651428,
+      "eval_runtime": 0.2048,
+      "eval_samples_per_second": 1352.477,
+      "eval_steps_per_second": 14.648,
       "step": 390
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6750902527075813,
-      "eval_loss": 0.6230459809303284,
-      "eval_runtime": 0.2124,
-      "eval_samples_per_second": 1304.362,
-      "eval_steps_per_second": 14.127,
+      "eval_accuracy": 0.6642599277978339,
+      "eval_loss": 0.6047545075416565,
+      "eval_runtime": 0.1975,
+      "eval_samples_per_second": 1402.222,
+      "eval_steps_per_second": 15.187,
       "step": 468
     },
     {
       "epoch": 6.41,
       "learning_rate": 0.00019099590723055935,
-      "loss": 0.6432,
+      "loss": 0.6366,
       "step": 500
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.6787003610108303,
-      "eval_loss": 0.610024094581604,
-      "eval_runtime": 0.2125,
-      "eval_samples_per_second": 1303.805,
-      "eval_steps_per_second": 14.121,
+      "eval_accuracy": 0.7111913357400722,
+      "eval_loss": 0.5592861771583557,
+      "eval_runtime": 0.1967,
+      "eval_samples_per_second": 1408.382,
+      "eval_steps_per_second": 15.253,
       "step": 546
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.6787003610108303,
-      "eval_loss": 0.6000346541404724,
-      "eval_runtime": 0.2026,
-      "eval_samples_per_second": 1367.422,
-      "eval_steps_per_second": 14.81,
+      "eval_accuracy": 0.703971119133574,
+      "eval_loss": 0.5585260987281799,
+      "eval_runtime": 0.2088,
+      "eval_samples_per_second": 1326.752,
+      "eval_steps_per_second": 14.369,
       "step": 624
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.6534296028880866,
-      "eval_loss": 0.6206708550453186,
-      "eval_runtime": 0.2041,
-      "eval_samples_per_second": 1357.114,
-      "eval_steps_per_second": 14.698,
+      "eval_accuracy": 0.7111913357400722,
+      "eval_loss": 0.571499764919281,
+      "eval_runtime": 0.2055,
+      "eval_samples_per_second": 1347.626,
+      "eval_steps_per_second": 14.595,
       "step": 702
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.6678700361010831,
-      "eval_loss": 0.6108516454696655,
-      "eval_runtime": 0.1973,
-      "eval_samples_per_second": 1404.203,
-      "eval_steps_per_second": 15.208,
+      "eval_accuracy": 0.7220216606498195,
+      "eval_loss": 0.5765948295593262,
+      "eval_runtime": 0.1974,
+      "eval_samples_per_second": 1403.091,
+      "eval_steps_per_second": 15.196,
       "step": 780
     },
     {
       "epoch": 10.0,
       "step": 780,
       "total_flos": 1638956800972800.0,
-      "train_loss": 0.5981983282627203,
-      "train_runtime": 104.8491,
-      "train_samples_per_second": 237.484,
-      "train_steps_per_second": 7.439
+      "train_loss": 0.5880965404021434,
+      "train_runtime": 74.518,
+      "train_samples_per_second": 334.147,
+      "train_steps_per_second": 10.467
     }
   ],
   "max_steps": 780,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dab0d68b1b046d6cb70008d1bc51e115e59f9d041f646d8215785aeef41b7265
+oid sha256:d942a12835e627c5832563e175e275a68f55113feef1e6c7484e8dc0409bd4fa
 size 2991