Training in progress, epoch 1

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.5270758122743683,
-    "eval_loss": 0.6929515600204468,
-    "eval_runtime": 0.3612,
+    "eval_accuracy": 0.6750902527075813,
+    "eval_loss": 0.6114864945411682,
+    "eval_runtime": 0.1984,
     "eval_samples": 277,
-    "eval_samples_per_second": 766.792,
-    "eval_steps_per_second": 8.305,
-    "train_loss": 0.6998463752942208,
-    "train_runtime": 85.2333,
+    "eval_samples_per_second": 1396.357,
+    "eval_steps_per_second": 15.123,
+    "train_loss": 0.5984805376101763,
+    "train_runtime": 104.4798,
     "train_samples": 2490,
-    "train_samples_per_second": 292.139,
-    "train_steps_per_second": 9.151
+    "train_samples_per_second": 238.323,
+    "train_steps_per_second": 7.466
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.5270758122743683,
-    "eval_loss": 0.6929515600204468,
-    "eval_runtime": 0.3612,
+    "eval_accuracy": 0.6750902527075813,
+    "eval_loss": 0.6114864945411682,
+    "eval_runtime": 0.1984,
     "eval_samples": 277,
-    "eval_samples_per_second": 766.792,
-    "eval_steps_per_second": 8.305
+    "eval_samples_per_second": 1396.357,
+    "eval_steps_per_second": 15.123
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2585f684674f08851e4654384c6fdbf4ee1ef3504d38065d1eb6d814f45af6c
-size 2661529
+oid sha256:186f3a00a6d3211008923fc2a8dcbca14f684341100ab01540c2dce27d7c6ef6
+size 7217721

runs/Jan29_13-50-41_node4/events.out.tfevents.1643435470.node4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26dd07379a7e286326e0d5667f4b61f4077db5310abd8e0a68a877e7817dcccf
-size 6251
+oid sha256:df050533b860a031db0f413cd4b099599803adefa890f186b2f44fa71a794a24
+size 7251

runs/Jan29_13-50-41_node4/events.out.tfevents.1643435575.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a1c97dae58fb7f88fc53b3b6d30fdaad465005395efd2c02a4956531759a3ec
+size 363

runs/Jan29_14-07-28_node4/1643436469.5333128/events.out.tfevents.1643436469.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbc57372fc5a9c0af60578d593ed8484e9152481c4d466fe91ed2346b38f0e00
+size 4585

runs/Jan29_14-07-28_node4/events.out.tfevents.1643436469.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:518d660a53601d42caa54afa6debbff60dbbd7b22036b41beb7f392fd54403b6
+size 3516

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 0.6998463752942208,
-    "train_runtime": 85.2333,
+    "train_loss": 0.5984805376101763,
+    "train_runtime": 104.4798,
     "train_samples": 2490,
-    "train_samples_per_second": 292.139,
-    "train_steps_per_second": 9.151
+    "train_samples_per_second": 238.323,
+    "train_steps_per_second": 7.466
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.5270758122743683,
-  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-312",
+  "best_metric": 0.6750902527075813,
+  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-624",
   "epoch": 10.0,
   "global_step": 780,
   "is_hyper_param_search": false,
@@ -10,112 +10,112 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.6938314437866211,
-      "eval_runtime": 0.2645,
-      "eval_samples_per_second": 1047.294,
-      "eval_steps_per_second": 11.343,
+      "eval_loss": 0.70159912109375,
+      "eval_runtime": 0.2363,
+      "eval_samples_per_second": 1172.313,
+      "eval_steps_per_second": 12.697,
       "step": 78
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.6956566572189331,
-      "eval_runtime": 0.1955,
-      "eval_samples_per_second": 1416.775,
-      "eval_steps_per_second": 15.344,
+      "eval_accuracy": 0.5018050541516246,
+      "eval_loss": 0.6921780109405518,
+      "eval_runtime": 0.1983,
+      "eval_samples_per_second": 1396.542,
+      "eval_steps_per_second": 15.125,
       "step": 156
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.6933903098106384,
-      "eval_runtime": 0.2248,
-      "eval_samples_per_second": 1232.274,
-      "eval_steps_per_second": 13.346,
+      "eval_accuracy": 0.5667870036101083,
+      "eval_loss": 0.6696296334266663,
+      "eval_runtime": 0.1952,
+      "eval_samples_per_second": 1419.099,
+      "eval_steps_per_second": 15.369,
       "step": 234
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.6929515600204468,
-      "eval_runtime": 0.1967,
-      "eval_samples_per_second": 1408.198,
-      "eval_steps_per_second": 15.251,
+      "eval_accuracy": 0.631768953068592,
+      "eval_loss": 0.643951416015625,
+      "eval_runtime": 0.2015,
+      "eval_samples_per_second": 1374.381,
+      "eval_steps_per_second": 14.885,
       "step": 312
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.7013436555862427,
-      "eval_runtime": 0.2215,
-      "eval_samples_per_second": 1250.465,
-      "eval_steps_per_second": 13.543,
+      "eval_accuracy": 0.631768953068592,
+      "eval_loss": 0.6657952070236206,
+      "eval_runtime": 0.1948,
+      "eval_samples_per_second": 1422.01,
+      "eval_steps_per_second": 15.401,
       "step": 390
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.5270758122743683,
-      "eval_loss": 0.6958906650543213,
-      "eval_runtime": 0.1949,
-      "eval_samples_per_second": 1420.918,
-      "eval_steps_per_second": 15.389,
+      "eval_accuracy": 0.6570397111913358,
+      "eval_loss": 0.6278102993965149,
+      "eval_runtime": 0.1944,
+      "eval_samples_per_second": 1424.734,
+      "eval_steps_per_second": 15.43,
       "step": 468
     },
     {
       "epoch": 6.41,
       "learning_rate": 0.00019099590723055935,
-      "loss": 0.7027,
+      "loss": 0.6432,
       "step": 500
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.6951103806495667,
-      "eval_runtime": 0.2035,
-      "eval_samples_per_second": 1360.881,
-      "eval_steps_per_second": 14.739,
+      "eval_accuracy": 0.6642599277978339,
+      "eval_loss": 0.6154282689094543,
+      "eval_runtime": 0.2257,
+      "eval_samples_per_second": 1227.515,
+      "eval_steps_per_second": 13.294,
       "step": 546
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.6951630711555481,
-      "eval_runtime": 0.1983,
-      "eval_samples_per_second": 1396.879,
-      "eval_steps_per_second": 15.129,
+      "eval_accuracy": 0.6750902527075813,
+      "eval_loss": 0.6114864945411682,
+      "eval_runtime": 0.2056,
+      "eval_samples_per_second": 1347.068,
+      "eval_steps_per_second": 14.589,
       "step": 624
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.6940933465957642,
-      "eval_runtime": 0.2,
-      "eval_samples_per_second": 1384.676,
-      "eval_steps_per_second": 14.996,
+      "eval_accuracy": 0.6534296028880866,
+      "eval_loss": 0.6326375603675842,
+      "eval_runtime": 0.1939,
+      "eval_samples_per_second": 1428.743,
+      "eval_steps_per_second": 15.474,
       "step": 702
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.694451630115509,
-      "eval_runtime": 0.2448,
-      "eval_samples_per_second": 1131.765,
-      "eval_steps_per_second": 12.257,
+      "eval_accuracy": 0.6678700361010831,
+      "eval_loss": 0.6225477457046509,
+      "eval_runtime": 0.2062,
+      "eval_samples_per_second": 1343.504,
+      "eval_steps_per_second": 14.551,
       "step": 780
     },
     {
       "epoch": 10.0,
       "step": 780,
-      "total_flos": 1660792129689600.0,
-      "train_loss": 0.6998463752942208,
-      "train_runtime": 85.2333,
-      "train_samples_per_second": 292.139,
-      "train_steps_per_second": 9.151
+      "total_flos": 1638956800972800.0,
+      "train_loss": 0.5984805376101763,
+      "train_runtime": 104.4798,
+      "train_samples_per_second": 238.323,
+      "train_steps_per_second": 7.466
     }
   ],
   "max_steps": 780,
   "num_train_epochs": 10,
-  "total_flos": 1660792129689600.0,
+  "total_flos": 1638956800972800.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:158f9c592b0108c1fc37ee53cdf7e163fa2c36d2fd97522f75d4bd5fc6611ad3
+oid sha256:a70b9f8e168ce5ea9504e84c452f8697d0356fdd9be2fde41b54fec4023942a1
 size 2991