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 MNLI
       type: glue
       config: mnli
       split: validation_matched
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.3544574630667346
+      value: 0.350793327908869
 ---
 
 <!-- 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. -->
 
 # hBERTv2_new_pretrain_48_mnli
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1002
-- Accuracy: 0.3545
+- Loss: 1.0885
+- Accuracy: 0.3508
 
 ## Model description
 

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 6.0,
+    "epoch_mm": 6.0,
+    "eval_accuracy": 0.34824248599083035,
+    "eval_accuracy_mm": 0.350793327908869,
+    "eval_loss": 1.0921986103057861,
+    "eval_loss_mm": 1.088494896888733,
+    "eval_runtime": 15.8132,
+    "eval_runtime_mm": 15.8435,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 620.683,
+    "eval_samples_per_second_mm": 620.572,
+    "eval_steps_per_second": 4.869,
+    "eval_steps_per_second_mm": 4.86,
+    "train_loss": 1.0960053166634411,
+    "train_runtime": 11010.3736,
+    "train_samples": 392702,
+    "train_samples_per_second": 1783.327,
+    "train_steps_per_second": 13.932
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 6.0,
+    "epoch_mm": 6.0,
+    "eval_accuracy": 0.34824248599083035,
+    "eval_accuracy_mm": 0.350793327908869,
+    "eval_loss": 1.0921986103057861,
+    "eval_loss_mm": 1.088494896888733,
+    "eval_runtime": 15.8132,
+    "eval_runtime_mm": 15.8435,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 620.683,
+    "eval_samples_per_second_mm": 620.572,
+    "eval_steps_per_second": 4.869,
+    "eval_steps_per_second_mm": 4.86
+}

logs/events.out.tfevents.1686066583.serv-9207.2387813.26

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55196fc965106336233fdab305a652cb707d69a1edd0d989ae8688869225a116
+size 698

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 1.0960053166634411,
+    "train_runtime": 11010.3736,
+    "train_samples": 392702,
+    "train_samples_per_second": 1783.327,
+    "train_steps_per_second": 13.932
+}

trainer_state.json

@@ -0,0 +1,115 @@
+{
+  "best_metric": 1.0921986103057861,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_48_mnli/checkpoint-3068",
+  "epoch": 6.0,
+  "global_step": 18408,
+  "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": 1.1,
+      "step": 3068
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.34824248599083035,
+      "eval_loss": 1.0921986103057861,
+      "eval_runtime": 16.2945,
+      "eval_samples_per_second": 602.349,
+      "eval_steps_per_second": 4.726,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 1.0893,
+      "step": 6136
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.3611818644931228,
+      "eval_loss": 1.096169352531433,
+      "eval_runtime": 16.027,
+      "eval_samples_per_second": 612.406,
+      "eval_steps_per_second": 4.804,
+      "step": 6136
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 3.76e-05,
+      "loss": 1.0894,
+      "step": 9204
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.3273560876209883,
+      "eval_loss": 1.0966593027114868,
+      "eval_runtime": 16.2826,
+      "eval_samples_per_second": 602.791,
+      "eval_steps_per_second": 4.729,
+      "step": 9204
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 1.0995,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.31818644931227713,
+      "eval_loss": 1.0970277786254883,
+      "eval_runtime": 16.2872,
+      "eval_samples_per_second": 602.621,
+      "eval_steps_per_second": 4.728,
+      "step": 12272
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.6e-05,
+      "loss": 1.0987,
+      "step": 15340
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.1009957790374756,
+      "eval_runtime": 16.05,
+      "eval_samples_per_second": 611.526,
+      "eval_steps_per_second": 4.798,
+      "step": 15340
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.52e-05,
+      "loss": 1.099,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.3544574630667346,
+      "eval_loss": 1.100155234336853,
+      "eval_runtime": 16.0117,
+      "eval_samples_per_second": 612.987,
+      "eval_steps_per_second": 4.809,
+      "step": 18408
+    },
+    {
+      "epoch": 6.0,
+      "step": 18408,
+      "total_flos": 3.448088908749865e+17,
+      "train_loss": 1.0960053166634411,
+      "train_runtime": 11010.3736,
+      "train_samples_per_second": 1783.327,
+      "train_steps_per_second": 13.932
+    }
+  ],
+  "max_steps": 153400,
+  "num_train_epochs": 50,
+  "total_flos": 3.448088908749865e+17,
+  "trial_name": null,
+  "trial_params": null
+}