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 COLA
       type: glue
       config: cola
       split: validation
@@ -28,9 +30,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hBERTv2_new_pretrain_cola
 
-This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new) on the glue dataset.
+This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new) on the GLUE COLA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 79.4784
+- Loss: 9.8472
 - Matthews Correlation: 0.0
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 6.0,
+    "eval_loss": 9.847195625305176,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.8224,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 572.309,
+    "eval_steps_per_second": 4.938,
+    "train_loss": 157.5859132074005,
+    "train_runtime": 312.4982,
+    "train_samples": 8551,
+    "train_samples_per_second": 1368.168,
+    "train_steps_per_second": 10.72
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 6.0,
+    "eval_loss": 9.847195625305176,
+    "eval_matthews_correlation": 0.0,
+    "eval_runtime": 1.8224,
+    "eval_samples": 1043,
+    "eval_samples_per_second": 572.309,
+    "eval_steps_per_second": 4.938
+}

logs/events.out.tfevents.1685526146.serv-3318.2595329.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2887c15be2ffdf3cb35cc42c121185b9a49d51ad710cff285aaad243e172e43d
+size 375

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "train_loss": 157.5859132074005,
+    "train_runtime": 312.4982,
+    "train_samples": 8551,
+    "train_samples_per_second": 1368.168,
+    "train_steps_per_second": 10.72
+}

trainer_state.json

@@ -0,0 +1,115 @@
+{
+  "best_metric": 9.847195625305176,
+  "best_model_checkpoint": "hBERTv2_new_pretrain_cola/checkpoint-67",
+  "epoch": 6.0,
+  "global_step": 402,
+  "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": 225.85,
+      "step": 67
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 9.847195625305176,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.8271,
+      "eval_samples_per_second": 570.863,
+      "eval_steps_per_second": 4.926,
+      "step": 67
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 322.5451,
+      "step": 134
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 176.96644592285156,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.8295,
+      "eval_samples_per_second": 570.11,
+      "eval_steps_per_second": 4.919,
+      "step": 134
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 88.5896,
+      "step": 201
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 127.87056732177734,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.8269,
+      "eval_samples_per_second": 570.919,
+      "eval_steps_per_second": 4.926,
+      "step": 201
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 121.8729,
+      "step": 268
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 106.45637512207031,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.8301,
+      "eval_samples_per_second": 569.928,
+      "eval_steps_per_second": 4.918,
+      "step": 268
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 99.3696,
+      "step": 335
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 85.71812438964844,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.827,
+      "eval_samples_per_second": 570.868,
+      "eval_steps_per_second": 4.926,
+      "step": 335
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 87.2883,
+      "step": 402
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 79.47842407226562,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 1.8262,
+      "eval_samples_per_second": 571.122,
+      "eval_steps_per_second": 4.928,
+      "step": 402
+    },
+    {
+      "epoch": 6.0,
+      "step": 402,
+      "total_flos": 7508077427294208.0,
+      "train_loss": 157.5859132074005,
+      "train_runtime": 312.4982,
+      "train_samples_per_second": 1368.168,
+      "train_steps_per_second": 10.72
+    }
+  ],
+  "max_steps": 3350,
+  "num_train_epochs": 50,
+  "total_flos": 7508077427294208.0,
+  "trial_name": null,
+  "trial_params": null
+}