End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE QNLI
       type: glue
       config: qnli
       split: validation
@@ -28,7 +30,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_no_pretrain_qnli
 
-This model is a fine-tuned version of [](https://huggingface.co/) on the glue dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6931
 - Accuracy: 0.5054

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 16.0,
+    "epoch": 13.0,
     "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.6930651664733887,
-    "eval_runtime": 9.4904,
+    "eval_loss": 0.6930924654006958,
+    "eval_runtime": 25.0729,
     "eval_samples": 5463,
-    "eval_samples_per_second": 575.636,
-    "eval_steps_per_second": 4.531,
-    "train_loss": 0.694585192334521,
-    "train_runtime": 8903.1517,
+    "eval_samples_per_second": 217.884,
+    "eval_steps_per_second": 2.273,
+    "train_loss": 0.6944669291952289,
+    "train_runtime": 19598.9141,
     "train_samples": 104743,
-    "train_samples_per_second": 588.236,
-    "train_steps_per_second": 4.599
+    "train_samples_per_second": 267.216,
+    "train_steps_per_second": 2.786
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 16.0,
+    "epoch": 13.0,
     "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.6930651664733887,
-    "eval_runtime": 9.4904,
+    "eval_loss": 0.6930924654006958,
+    "eval_runtime": 25.0729,
     "eval_samples": 5463,
-    "eval_samples_per_second": 575.636,
-    "eval_steps_per_second": 4.531
+    "eval_samples_per_second": 217.884,
+    "eval_steps_per_second": 2.273
 }

logs/events.out.tfevents.1686778395.garda.2667877.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30e4ddc7c170986810ade6c5ceba689a63deb9f97079f78735ba4caa997d28a1
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 16.0,
-    "train_loss": 0.694585192334521,
-    "train_runtime": 8903.1517,
+    "epoch": 13.0,
+    "train_loss": 0.6944669291952289,
+    "train_runtime": 19598.9141,
     "train_samples": 104743,
-    "train_samples_per_second": 588.236,
-    "train_steps_per_second": 4.599
+    "train_samples_per_second": 267.216,
+    "train_steps_per_second": 2.786
 }

trainer_state.json

@@ -1,265 +1,220 @@
 {
-  "best_metric": 0.6930651664733887,
-  "best_model_checkpoint": "hBERTv1_no_pretrain_qnli/checkpoint-9009",
-  "epoch": 16.0,
-  "global_step": 13104,
+  "best_metric": 0.6930924654006958,
+  "best_model_checkpoint": "hBERTv1_no_pretrain_qnli/checkpoint-8736",
+  "epoch": 13.0,
+  "global_step": 14196,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.00049002442002442,
-      "loss": 0.715,
-      "step": 819
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.7059,
+      "step": 1092
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6931389570236206,
-      "eval_runtime": 9.5675,
-      "eval_samples_per_second": 570.995,
-      "eval_steps_per_second": 4.494,
-      "step": 819
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.7003825306892395,
+      "eval_runtime": 25.3252,
+      "eval_samples_per_second": 215.714,
+      "eval_steps_per_second": 2.251,
+      "step": 1092
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048002442002442004,
-      "loss": 0.6932,
-      "step": 1638
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6948,
+      "step": 2184
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6931495070457458,
-      "eval_runtime": 9.5694,
-      "eval_samples_per_second": 570.88,
-      "eval_steps_per_second": 4.493,
-      "step": 1638
+      "eval_loss": 0.693818986415863,
+      "eval_runtime": 25.5889,
+      "eval_samples_per_second": 213.491,
+      "eval_steps_per_second": 2.228,
+      "step": 2184
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047002442002442,
-      "loss": 0.6936,
-      "step": 2457
+      "learning_rate": 3.76e-05,
+      "loss": 0.6939,
+      "step": 3276
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6931167840957642,
-      "eval_runtime": 9.5644,
-      "eval_samples_per_second": 571.18,
-      "eval_steps_per_second": 4.496,
-      "step": 2457
+      "eval_loss": 0.6931672692298889,
+      "eval_runtime": 25.592,
+      "eval_samples_per_second": 213.465,
+      "eval_steps_per_second": 2.227,
+      "step": 3276
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046002442002442004,
-      "loss": 0.6932,
-      "step": 3276
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6936,
+      "step": 4368
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6931969523429871,
-      "eval_runtime": 9.5713,
-      "eval_samples_per_second": 570.77,
-      "eval_steps_per_second": 4.493,
-      "step": 3276
+      "eval_accuracy": 0.5053999633900788,
+      "eval_loss": 0.693107545375824,
+      "eval_runtime": 25.5969,
+      "eval_samples_per_second": 213.424,
+      "eval_steps_per_second": 2.227,
+      "step": 4368
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045002442002442,
-      "loss": 0.6932,
-      "step": 4095
+      "learning_rate": 3.6e-05,
+      "loss": 0.6934,
+      "step": 5460
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933488249778748,
-      "eval_runtime": 9.573,
-      "eval_samples_per_second": 570.669,
-      "eval_steps_per_second": 4.492,
-      "step": 4095
+      "eval_loss": 0.6930991411209106,
+      "eval_runtime": 25.4035,
+      "eval_samples_per_second": 215.05,
+      "eval_steps_per_second": 2.244,
+      "step": 5460
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044002442002442004,
-      "loss": 0.6932,
-      "step": 4914
+      "learning_rate": 3.52e-05,
+      "loss": 0.6936,
+      "step": 6552
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6930915117263794,
-      "eval_runtime": 9.5567,
-      "eval_samples_per_second": 571.639,
-      "eval_steps_per_second": 4.499,
-      "step": 4914
+      "eval_loss": 0.6931207180023193,
+      "eval_runtime": 25.4067,
+      "eval_samples_per_second": 215.022,
+      "eval_steps_per_second": 2.244,
+      "step": 6552
     },
     {
       "epoch": 7.0,
-      "learning_rate": 0.00043002442002442007,
-      "loss": 0.6932,
-      "step": 5733
+      "learning_rate": 3.44e-05,
+      "loss": 0.6933,
+      "step": 7644
     },
     {
       "epoch": 7.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6931073069572449,
-      "eval_runtime": 9.5808,
-      "eval_samples_per_second": 570.202,
-      "eval_steps_per_second": 4.488,
-      "step": 5733
+      "eval_loss": 0.693096935749054,
+      "eval_runtime": 25.4153,
+      "eval_samples_per_second": 214.949,
+      "eval_steps_per_second": 2.243,
+      "step": 7644
     },
     {
       "epoch": 8.0,
-      "learning_rate": 0.00042002442002442005,
-      "loss": 0.6932,
-      "step": 6552
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6933,
+      "step": 8736
     },
     {
       "epoch": 8.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6930862069129944,
-      "eval_runtime": 9.5904,
-      "eval_samples_per_second": 569.63,
-      "eval_steps_per_second": 4.484,
-      "step": 6552
+      "eval_loss": 0.6930924654006958,
+      "eval_runtime": 25.5577,
+      "eval_samples_per_second": 213.751,
+      "eval_steps_per_second": 2.23,
+      "step": 8736
     },
     {
       "epoch": 9.0,
-      "learning_rate": 0.00041003663003663003,
-      "loss": 0.6935,
-      "step": 7371
+      "learning_rate": 3.28e-05,
+      "loss": 0.6934,
+      "step": 9828
     },
     {
       "epoch": 9.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.693478524684906,
-      "eval_runtime": 9.5344,
-      "eval_samples_per_second": 572.981,
-      "eval_steps_per_second": 4.51,
-      "step": 7371
+      "eval_loss": 0.6934182047843933,
+      "eval_runtime": 25.5166,
+      "eval_samples_per_second": 214.096,
+      "eval_steps_per_second": 2.234,
+      "step": 9828
     },
     {
       "epoch": 10.0,
-      "learning_rate": 0.00040003663003663006,
-      "loss": 0.6932,
-      "step": 8190
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.6933,
+      "step": 10920
     },
     {
       "epoch": 10.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6930915117263794,
-      "eval_runtime": 9.5321,
-      "eval_samples_per_second": 573.116,
-      "eval_steps_per_second": 4.511,
-      "step": 8190
+      "eval_loss": 0.6930983662605286,
+      "eval_runtime": 25.3329,
+      "eval_samples_per_second": 215.649,
+      "eval_steps_per_second": 2.25,
+      "step": 10920
     },
     {
       "epoch": 11.0,
-      "learning_rate": 0.00039003663003663004,
+      "learning_rate": 3.1200000000000006e-05,
       "loss": 0.6932,
-      "step": 9009
+      "step": 12012
     },
     {
       "epoch": 11.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6930651664733887,
-      "eval_runtime": 9.5448,
-      "eval_samples_per_second": 572.352,
-      "eval_steps_per_second": 4.505,
-      "step": 9009
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6932870149612427,
+      "eval_runtime": 25.3051,
+      "eval_samples_per_second": 215.885,
+      "eval_steps_per_second": 2.253,
+      "step": 12012
     },
     {
       "epoch": 12.0,
-      "learning_rate": 0.00038004884004884,
+      "learning_rate": 3.0400000000000004e-05,
       "loss": 0.6932,
-      "step": 9828
+      "step": 13104
     },
     {
       "epoch": 12.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6931073069572449,
-      "eval_runtime": 9.5697,
-      "eval_samples_per_second": 570.866,
-      "eval_steps_per_second": 4.493,
-      "step": 9828
+      "eval_loss": 0.6931389570236206,
+      "eval_runtime": 25.2541,
+      "eval_samples_per_second": 216.322,
+      "eval_steps_per_second": 2.257,
+      "step": 13104
     },
     {
       "epoch": 13.0,
-      "learning_rate": 0.00037004884004884005,
-      "loss": 0.6932,
-      "step": 10647
+      "learning_rate": 2.96e-05,
+      "loss": 0.6933,
+      "step": 14196
     },
     {
       "epoch": 13.0,
       "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6931073069572449,
-      "eval_runtime": 9.5909,
-      "eval_samples_per_second": 569.603,
-      "eval_steps_per_second": 4.483,
-      "step": 10647
-    },
-    {
-      "epoch": 14.0,
-      "learning_rate": 0.00036004884004884,
-      "loss": 0.6932,
-      "step": 11466
+      "eval_loss": 0.6931240558624268,
+      "eval_runtime": 25.2965,
+      "eval_samples_per_second": 215.959,
+      "eval_steps_per_second": 2.253,
+      "step": 14196
     },
     {
-      "epoch": 14.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6931231617927551,
-      "eval_runtime": 9.5446,
-      "eval_samples_per_second": 572.363,
-      "eval_steps_per_second": 4.505,
-      "step": 11466
-    },
-    {
-      "epoch": 15.0,
-      "learning_rate": 0.00035004884004884005,
-      "loss": 0.6932,
-      "step": 12285
-    },
-    {
-      "epoch": 15.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.693396270275116,
-      "eval_runtime": 9.5573,
-      "eval_samples_per_second": 571.603,
-      "eval_steps_per_second": 4.499,
-      "step": 12285
-    },
-    {
-      "epoch": 16.0,
-      "learning_rate": 0.0003400488400488401,
-      "loss": 0.6932,
-      "step": 13104
-    },
-    {
-      "epoch": 16.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6931284070014954,
-      "eval_runtime": 9.5555,
-      "eval_samples_per_second": 571.71,
-      "eval_steps_per_second": 4.5,
-      "step": 13104
-    },
-    {
-      "epoch": 16.0,
-      "step": 13104,
-      "total_flos": 2.481936015514665e+17,
-      "train_loss": 0.694585192334521,
-      "train_runtime": 8903.1517,
-      "train_samples_per_second": 588.236,
-      "train_steps_per_second": 4.599
+      "epoch": 13.0,
+      "step": 14196,
+      "total_flos": 2.0165730126056653e+17,
+      "train_loss": 0.6944669291952289,
+      "train_runtime": 19598.9141,
+      "train_samples_per_second": 267.216,
+      "train_steps_per_second": 2.786
     }
   ],
-  "max_steps": 40950,
+  "max_steps": 54600,
   "num_train_epochs": 50,
-  "total_flos": 2.481936015514665e+17,
+  "total_flos": 2.0165730126056653e+17,
   "trial_name": null,
   "trial_params": null
 }