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 RTE
       type: glue
       config: rte
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5379061371841155
+      value: 0.5306859205776173
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv1_new_pretrain_rte
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new) on the GLUE RTE dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2096
-- Accuracy: 0.5379
+- Loss: 0.6896
+- Accuracy: 0.5307
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.4729241877256318,
-    "eval_loss": 28.268051147460938,
-    "eval_runtime": 0.5114,
+    "epoch": 12.0,
+    "eval_accuracy": 0.5306859205776173,
+    "eval_loss": 0.6895588636398315,
+    "eval_runtime": 0.511,
     "eval_samples": 277,
-    "eval_samples_per_second": 541.669,
-    "eval_steps_per_second": 5.866,
-    "train_loss": 29.028575642903647,
-    "train_runtime": 102.0364,
+    "eval_samples_per_second": 542.108,
+    "eval_steps_per_second": 5.871,
+    "train_loss": 0.6050535062948863,
+    "train_runtime": 211.7637,
     "train_samples": 2490,
-    "train_samples_per_second": 1220.153,
-    "train_steps_per_second": 9.8
+    "train_samples_per_second": 587.919,
+    "train_steps_per_second": 4.722
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 6.0,
-    "eval_accuracy": 0.4729241877256318,
-    "eval_loss": 28.268051147460938,
-    "eval_runtime": 0.5114,
+    "epoch": 12.0,
+    "eval_accuracy": 0.5306859205776173,
+    "eval_loss": 0.6895588636398315,
+    "eval_runtime": 0.511,
     "eval_samples": 277,
-    "eval_samples_per_second": 541.669,
-    "eval_steps_per_second": 5.866
+    "eval_samples_per_second": 542.108,
+    "eval_steps_per_second": 5.871
 }

logs/events.out.tfevents.1686053142.serv-3317.3374904.17

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2426ed4f7f615cd75f586c3f6577612bc7d6af0261215733064caaddc34018e
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 6.0,
-    "train_loss": 29.028575642903647,
-    "train_runtime": 102.0364,
+    "epoch": 12.0,
+    "train_loss": 0.6050535062948863,
+    "train_runtime": 211.7637,
     "train_samples": 2490,
-    "train_samples_per_second": 1220.153,
-    "train_steps_per_second": 9.8
+    "train_samples_per_second": 587.919,
+    "train_steps_per_second": 4.722
 }

trainer_state.json

@@ -1,115 +1,205 @@
 {
-  "best_metric": 28.268051147460938,
-  "best_model_checkpoint": "hBERTv1_new_pretrain_rte/checkpoint-20",
-  "epoch": 6.0,
-  "global_step": 120,
+  "best_metric": 0.6895588636398315,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_rte/checkpoint-140",
+  "epoch": 12.0,
+  "global_step": 240,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.00049,
-      "loss": 28.8034,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.7407,
       "step": 20
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 28.268051147460938,
-      "eval_runtime": 0.5081,
-      "eval_samples_per_second": 545.128,
-      "eval_steps_per_second": 5.904,
+      "eval_loss": 0.7002341151237488,
+      "eval_runtime": 0.5049,
+      "eval_samples_per_second": 548.575,
+      "eval_steps_per_second": 5.941,
       "step": 20
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048,
-      "loss": 28.8444,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.7061,
       "step": 40
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 28.268051147460938,
-      "eval_runtime": 0.5097,
-      "eval_samples_per_second": 543.476,
-      "eval_steps_per_second": 5.886,
+      "eval_loss": 0.7245458960533142,
+      "eval_runtime": 0.508,
+      "eval_samples_per_second": 545.223,
+      "eval_steps_per_second": 5.905,
       "step": 40
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047,
-      "loss": 28.6738,
+      "learning_rate": 3.76e-05,
+      "loss": 0.7102,
       "step": 60
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 28.268051147460938,
-      "eval_runtime": 0.5098,
-      "eval_samples_per_second": 543.347,
-      "eval_steps_per_second": 5.885,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6948612332344055,
+      "eval_runtime": 0.504,
+      "eval_samples_per_second": 549.561,
+      "eval_steps_per_second": 5.952,
       "step": 60
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046,
-      "loss": 29.3662,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.703,
       "step": 80
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 28.268051147460938,
-      "eval_runtime": 0.5103,
-      "eval_samples_per_second": 542.814,
-      "eval_steps_per_second": 5.879,
+      "eval_loss": 0.6950586438179016,
+      "eval_runtime": 0.5087,
+      "eval_samples_per_second": 544.51,
+      "eval_steps_per_second": 5.897,
       "step": 80
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045000000000000004,
-      "loss": 30.0413,
+      "learning_rate": 3.6e-05,
+      "loss": 0.7097,
       "step": 100
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 28.268051147460938,
-      "eval_runtime": 0.5111,
-      "eval_samples_per_second": 541.935,
-      "eval_steps_per_second": 5.869,
+      "eval_loss": 0.6973713636398315,
+      "eval_runtime": 0.5068,
+      "eval_samples_per_second": 546.522,
+      "eval_steps_per_second": 5.919,
       "step": 100
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044,
-      "loss": 28.4423,
+      "learning_rate": 3.52e-05,
+      "loss": 0.7006,
       "step": 120
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 28.268051147460938,
-      "eval_runtime": 0.5109,
-      "eval_samples_per_second": 542.142,
-      "eval_steps_per_second": 5.872,
+      "eval_loss": 0.7052543759346008,
+      "eval_runtime": 0.5068,
+      "eval_samples_per_second": 546.614,
+      "eval_steps_per_second": 5.92,
       "step": 120
     },
     {
-      "epoch": 6.0,
-      "step": 120,
-      "total_flos": 2212565757198336.0,
-      "train_loss": 29.028575642903647,
-      "train_runtime": 102.0364,
-      "train_samples_per_second": 1220.153,
-      "train_steps_per_second": 9.8
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6986,
+      "step": 140
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5306859205776173,
+      "eval_loss": 0.6895588636398315,
+      "eval_runtime": 0.5057,
+      "eval_samples_per_second": 547.783,
+      "eval_steps_per_second": 5.933,
+      "step": 140
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6935,
+      "step": 160
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.7711036205291748,
+      "eval_runtime": 0.5097,
+      "eval_samples_per_second": 543.49,
+      "eval_steps_per_second": 5.886,
+      "step": 160
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.6109,
+      "step": 180
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.4981949458483754,
+      "eval_loss": 0.8443405032157898,
+      "eval_runtime": 0.5053,
+      "eval_samples_per_second": 548.178,
+      "eval_steps_per_second": 5.937,
+      "step": 180
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.469,
+      "step": 200
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5126353790613718,
+      "eval_loss": 1.0369467735290527,
+      "eval_runtime": 0.5078,
+      "eval_samples_per_second": 545.529,
+      "eval_steps_per_second": 5.908,
+      "step": 200
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.3028,
+      "step": 220
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5234657039711191,
+      "eval_loss": 1.1621102094650269,
+      "eval_runtime": 0.512,
+      "eval_samples_per_second": 541.059,
+      "eval_steps_per_second": 5.86,
+      "step": 220
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.2155,
+      "step": 240
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5379061371841155,
+      "eval_loss": 1.2095954418182373,
+      "eval_runtime": 0.5057,
+      "eval_samples_per_second": 547.72,
+      "eval_steps_per_second": 5.932,
+      "step": 240
+    },
+    {
+      "epoch": 12.0,
+      "step": 240,
+      "total_flos": 4425131514396672.0,
+      "train_loss": 0.6050535062948863,
+      "train_runtime": 211.7637,
+      "train_samples_per_second": 587.919,
+      "train_steps_per_second": 4.722
     }
   ],
   "max_steps": 1000,
   "num_train_epochs": 50,
-  "total_flos": 2212565757198336.0,
+  "total_flos": 4425131514396672.0,
   "trial_name": null,
   "trial_params": null
 }