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: 0.5319421563243639
+      value: 0.5416437854658612
 ---
 
 <!-- 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. -->
 
 # add_BERT_24_qnli
 
-This model is a fine-tuned version of [gokuls/add_bert_12_layer_model_complete_training_new](https://huggingface.co/gokuls/add_bert_12_layer_model_complete_training_new) on the glue dataset.
+This model is a fine-tuned version of [gokuls/add_bert_12_layer_model_complete_training_new](https://huggingface.co/gokuls/add_bert_12_layer_model_complete_training_new) on the GLUE QNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6896
-- Accuracy: 0.5319
+- Loss: 0.6883
+- Accuracy: 0.5416
 
 ## Model description
 

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5416437854658612,
+    "eval_loss": 0.688339114189148,
+    "eval_runtime": 10.3508,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 527.788,
+    "eval_steps_per_second": 4.154,
+    "train_loss": 0.6927911774059584,
+    "train_runtime": 8972.1935,
+    "train_samples": 104743,
+    "train_samples_per_second": 583.709,
+    "train_steps_per_second": 4.564
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 14.0,
+    "eval_accuracy": 0.5416437854658612,
+    "eval_loss": 0.688339114189148,
+    "eval_runtime": 10.3508,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 527.788,
+    "eval_steps_per_second": 4.154
+}

logs/events.out.tfevents.1686873012.serv-3317.1852327.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f773db18973c36784282b8c189163aa4fdb5209ad92660d97877525487db256
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 14.0,
+    "train_loss": 0.6927911774059584,
+    "train_runtime": 8972.1935,
+    "train_samples": 104743,
+    "train_samples_per_second": 583.709,
+    "train_steps_per_second": 4.564
+}

trainer_state.json

@@ -0,0 +1,235 @@
+{
+  "best_metric": 0.688339114189148,
+  "best_model_checkpoint": "add_BERT_24_qnli/checkpoint-7371",
+  "epoch": 14.0,
+  "global_step": 11466,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 0.6964,
+      "step": 819
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5171151382024529,
+      "eval_loss": 0.6927340030670166,
+      "eval_runtime": 10.4326,
+      "eval_samples_per_second": 523.649,
+      "eval_steps_per_second": 4.122,
+      "step": 819
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 0.6946,
+      "step": 1638
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6926895976066589,
+      "eval_runtime": 10.4091,
+      "eval_samples_per_second": 524.828,
+      "eval_steps_per_second": 4.131,
+      "step": 1638
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 0.6936,
+      "step": 2457
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5200439319055464,
+      "eval_loss": 0.6915343999862671,
+      "eval_runtime": 10.4082,
+      "eval_samples_per_second": 524.877,
+      "eval_steps_per_second": 4.131,
+      "step": 2457
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 0.6998,
+      "step": 3276
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.4946000366099213,
+      "eval_loss": 0.6902428865432739,
+      "eval_runtime": 10.3971,
+      "eval_samples_per_second": 525.437,
+      "eval_steps_per_second": 4.136,
+      "step": 3276
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 0.6925,
+      "step": 4095
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5257184697052901,
+      "eval_loss": 0.6933450102806091,
+      "eval_runtime": 10.4062,
+      "eval_samples_per_second": 524.976,
+      "eval_steps_per_second": 4.132,
+      "step": 4095
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 0.6917,
+      "step": 4914
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5273659161632802,
+      "eval_loss": 0.6893199682235718,
+      "eval_runtime": 10.3963,
+      "eval_samples_per_second": 525.473,
+      "eval_steps_per_second": 4.136,
+      "step": 4914
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 3.44e-05,
+      "loss": 0.6914,
+      "step": 5733
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5293794618341571,
+      "eval_loss": 0.6894359588623047,
+      "eval_runtime": 10.4141,
+      "eval_samples_per_second": 524.579,
+      "eval_steps_per_second": 4.129,
+      "step": 5733
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 0.6916,
+      "step": 6552
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5381658429434377,
+      "eval_loss": 0.6887709498405457,
+      "eval_runtime": 10.4031,
+      "eval_samples_per_second": 525.134,
+      "eval_steps_per_second": 4.133,
+      "step": 6552
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 3.28e-05,
+      "loss": 0.6913,
+      "step": 7371
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5416437854658612,
+      "eval_loss": 0.688339114189148,
+      "eval_runtime": 10.416,
+      "eval_samples_per_second": 524.484,
+      "eval_steps_per_second": 4.128,
+      "step": 7371
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.6909,
+      "step": 8190
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5356031484532309,
+      "eval_loss": 0.6892455816268921,
+      "eval_runtime": 10.4058,
+      "eval_samples_per_second": 524.995,
+      "eval_steps_per_second": 4.132,
+      "step": 8190
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 0.6914,
+      "step": 9009
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.5410946366465312,
+      "eval_loss": 0.6891716718673706,
+      "eval_runtime": 10.3996,
+      "eval_samples_per_second": 525.308,
+      "eval_steps_per_second": 4.135,
+      "step": 9009
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 3.0400000000000004e-05,
+      "loss": 0.6918,
+      "step": 9828
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.5257184697052901,
+      "eval_loss": 0.6906862854957581,
+      "eval_runtime": 10.3858,
+      "eval_samples_per_second": 526.006,
+      "eval_steps_per_second": 4.14,
+      "step": 9828
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 2.96e-05,
+      "loss": 0.6911,
+      "step": 10647
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.5286472634083836,
+      "eval_loss": 0.6905463933944702,
+      "eval_runtime": 10.4034,
+      "eval_samples_per_second": 525.118,
+      "eval_steps_per_second": 4.133,
+      "step": 10647
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 2.8800000000000002e-05,
+      "loss": 0.6909,
+      "step": 11466
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.5319421563243639,
+      "eval_loss": 0.689621090888977,
+      "eval_runtime": 10.4366,
+      "eval_samples_per_second": 523.444,
+      "eval_steps_per_second": 4.12,
+      "step": 11466
+    },
+    {
+      "epoch": 14.0,
+      "step": 11466,
+      "total_flos": 2.0943735856562176e+17,
+      "train_loss": 0.6927911774059584,
+      "train_runtime": 8972.1935,
+      "train_samples_per_second": 583.709,
+      "train_steps_per_second": 4.564
+    }
+  ],
+  "max_steps": 40950,
+  "num_train_epochs": 50,
+  "total_flos": 2.0943735856562176e+17,
+  "trial_name": null,
+  "trial_params": null
+}