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
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.5427420831045213
+      value: 0.598572213069742
 ---
 
 <!-- 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_w_init__qnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v1_complete_training_new_wt_init](https://huggingface.co/gokuls/bert_12_layer_model_v1_complete_training_new_wt_init) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6908
-- Accuracy: 0.5427
+- Loss: 0.6672
+- Accuracy: 0.5986
 
 ## Model description
 

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 8.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.691364049911499,
-    "eval_runtime": 9.8448,
+    "epoch": 11.0,
+    "eval_accuracy": 0.598572213069742,
+    "eval_loss": 0.6672190427780151,
+    "eval_runtime": 9.7744,
     "eval_samples": 5463,
-    "eval_samples_per_second": 554.912,
-    "eval_steps_per_second": 4.368,
-    "train_loss": 0.7089429955488306,
-    "train_runtime": 4466.842,
+    "eval_samples_per_second": 558.909,
+    "eval_steps_per_second": 4.399,
+    "train_loss": 0.6730744434601856,
+    "train_runtime": 6372.3482,
     "train_samples": 104743,
-    "train_samples_per_second": 1172.45,
-    "train_steps_per_second": 9.168
+    "train_samples_per_second": 821.856,
+    "train_steps_per_second": 6.426
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 8.0,
-    "eval_accuracy": 0.5053999633900788,
-    "eval_loss": 0.691364049911499,
-    "eval_runtime": 9.8448,
+    "epoch": 11.0,
+    "eval_accuracy": 0.598572213069742,
+    "eval_loss": 0.6672190427780151,
+    "eval_runtime": 9.7744,
     "eval_samples": 5463,
-    "eval_samples_per_second": 554.912,
-    "eval_steps_per_second": 4.368
+    "eval_samples_per_second": 558.909,
+    "eval_steps_per_second": 4.399
 }

logs/events.out.tfevents.1686040226.serv-3317.3377907.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0788de25ebb62c9ec6570c462ebb3156a14807333e68d5a5d5c0da8eb902eb5
+size 363

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.0,
-    "train_loss": 0.7089429955488306,
-    "train_runtime": 4466.842,
+    "epoch": 11.0,
+    "train_loss": 0.6730744434601856,
+    "train_runtime": 6372.3482,
     "train_samples": 104743,
-    "train_samples_per_second": 1172.45,
-    "train_steps_per_second": 9.168
+    "train_samples_per_second": 821.856,
+    "train_steps_per_second": 6.426
 }

trainer_state.json

@@ -1,145 +1,190 @@
 {
-  "best_metric": 0.691364049911499,
-  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__qnli/checkpoint-2457",
-  "epoch": 8.0,
-  "global_step": 6552,
+  "best_metric": 0.6672190427780151,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__qnli/checkpoint-4914",
+  "epoch": 11.0,
+  "global_step": 9009,
   "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": 0.8188,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.6909,
       "step": 819
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6932117342948914,
-      "eval_runtime": 9.863,
-      "eval_samples_per_second": 553.888,
-      "eval_steps_per_second": 4.36,
+      "eval_accuracy": 0.5652571846970529,
+      "eval_loss": 0.6782859563827515,
+      "eval_runtime": 9.8372,
+      "eval_samples_per_second": 555.34,
+      "eval_steps_per_second": 4.371,
       "step": 819
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.00048,
-      "loss": 0.6937,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.684,
       "step": 1638
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6934226751327515,
-      "eval_runtime": 9.8555,
-      "eval_samples_per_second": 554.308,
-      "eval_steps_per_second": 4.363,
+      "eval_accuracy": 0.5099762035511624,
+      "eval_loss": 0.6904051899909973,
+      "eval_runtime": 9.816,
+      "eval_samples_per_second": 556.542,
+      "eval_steps_per_second": 4.381,
       "step": 1638
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.00047,
-      "loss": 0.6932,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6765,
       "step": 2457
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.691364049911499,
-      "eval_runtime": 9.8848,
-      "eval_samples_per_second": 552.667,
-      "eval_steps_per_second": 4.35,
+      "eval_accuracy": 0.5881383855024712,
+      "eval_loss": 0.6709150671958923,
+      "eval_runtime": 9.8166,
+      "eval_samples_per_second": 556.504,
+      "eval_steps_per_second": 4.38,
       "step": 2457
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046,
-      "loss": 0.6932,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6696,
       "step": 3276
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.4946000366099213,
-      "eval_loss": 0.6934226751327515,
-      "eval_runtime": 9.8572,
-      "eval_samples_per_second": 554.216,
-      "eval_steps_per_second": 4.362,
+      "eval_accuracy": 0.5694673256452498,
+      "eval_loss": 0.6773616671562195,
+      "eval_runtime": 9.8364,
+      "eval_samples_per_second": 555.383,
+      "eval_steps_per_second": 4.371,
       "step": 3276
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.00045000000000000004,
-      "loss": 0.6932,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6676,
       "step": 4095
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 9.8568,
-      "eval_samples_per_second": 554.238,
-      "eval_steps_per_second": 4.362,
+      "eval_accuracy": 0.5903349807797913,
+      "eval_loss": 0.6704134345054626,
+      "eval_runtime": 9.8333,
+      "eval_samples_per_second": 555.562,
+      "eval_steps_per_second": 4.373,
       "step": 4095
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044,
-      "loss": 0.6932,
+      "learning_rate": 3.52e-05,
+      "loss": 0.6626,
       "step": 4914
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 9.8513,
-      "eval_samples_per_second": 554.547,
-      "eval_steps_per_second": 4.365,
+      "eval_accuracy": 0.598572213069742,
+      "eval_loss": 0.6672190427780151,
+      "eval_runtime": 9.8615,
+      "eval_samples_per_second": 553.973,
+      "eval_steps_per_second": 4.36,
       "step": 4914
     },
     {
       "epoch": 7.0,
-      "learning_rate": 0.00043,
-      "loss": 0.6932,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6661,
       "step": 5733
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 9.9161,
-      "eval_samples_per_second": 550.923,
-      "eval_steps_per_second": 4.336,
+      "eval_accuracy": 0.5907010799926781,
+      "eval_loss": 0.6703054904937744,
+      "eval_runtime": 9.8372,
+      "eval_samples_per_second": 555.339,
+      "eval_steps_per_second": 4.371,
       "step": 5733
     },
     {
       "epoch": 8.0,
-      "learning_rate": 0.00042,
-      "loss": 0.6931,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6642,
       "step": 6552
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.5053999633900788,
-      "eval_loss": 0.6933382749557495,
-      "eval_runtime": 9.9544,
-      "eval_samples_per_second": 548.802,
-      "eval_steps_per_second": 4.32,
+      "eval_accuracy": 0.5960095185795351,
+      "eval_loss": 0.6692940592765808,
+      "eval_runtime": 9.8258,
+      "eval_samples_per_second": 555.985,
+      "eval_steps_per_second": 4.376,
       "step": 6552
     },
     {
-      "epoch": 8.0,
-      "step": 6552,
-      "total_flos": 1.2409680077573325e+17,
-      "train_loss": 0.7089429955488306,
-      "train_runtime": 4466.842,
-      "train_samples_per_second": 1172.45,
-      "train_steps_per_second": 9.168
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.6698,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5799011532125206,
+      "eval_loss": 0.6733358502388,
+      "eval_runtime": 9.846,
+      "eval_samples_per_second": 554.842,
+      "eval_steps_per_second": 4.367,
+      "step": 7371
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.6724,
+      "step": 8190
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5636097382390628,
+      "eval_loss": 0.6815194487571716,
+      "eval_runtime": 9.8244,
+      "eval_samples_per_second": 556.067,
+      "eval_steps_per_second": 4.377,
+      "step": 8190
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.68,
+      "step": 9009
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5427420831045213,
+      "eval_loss": 0.6908370852470398,
+      "eval_runtime": 9.8206,
+      "eval_samples_per_second": 556.277,
+      "eval_steps_per_second": 4.379,
+      "step": 9009
+    },
+    {
+      "epoch": 11.0,
+      "step": 9009,
+      "total_flos": 1.706331010666332e+17,
+      "train_loss": 0.6730744434601856,
+      "train_runtime": 6372.3482,
+      "train_samples_per_second": 821.856,
+      "train_steps_per_second": 6.426
     }
   ],
   "max_steps": 40950,
   "num_train_epochs": 50,
-  "total_flos": 1.2409680077573325e+17,
+  "total_flos": 1.706331010666332e+17,
   "trial_name": null,
   "trial_params": null
 }