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
@@ -28,9 +30,9 @@ 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: 0.6933
+- Loss: 0.6914
 - Accuracy: 0.5054
 
 ## Model description

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.691364049911499,
+    "eval_runtime": 9.8448,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 554.912,
+    "eval_steps_per_second": 4.368,
+    "train_loss": 0.7089429955488306,
+    "train_runtime": 4466.842,
+    "train_samples": 104743,
+    "train_samples_per_second": 1172.45,
+    "train_steps_per_second": 9.168
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.5053999633900788,
+    "eval_loss": 0.691364049911499,
+    "eval_runtime": 9.8448,
+    "eval_samples": 5463,
+    "eval_samples_per_second": 554.912,
+    "eval_steps_per_second": 4.368
+}

logs/events.out.tfevents.1685533545.serv-9223.3032138.11

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a6de94d4fcfac5baf84bb5a5510afd83b079b7c883e9867db6efa8f032b1452
+size 363

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 0.7089429955488306,
+    "train_runtime": 4466.842,
+    "train_samples": 104743,
+    "train_samples_per_second": 1172.45,
+    "train_steps_per_second": 9.168
+}

trainer_state.json

@@ -0,0 +1,145 @@
+{
+  "best_metric": 0.691364049911499,
+  "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__qnli/checkpoint-2457",
+  "epoch": 8.0,
+  "global_step": 6552,
+  "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,
+      "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,
+      "step": 819
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00048,
+      "loss": 0.6937,
+      "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,
+      "step": 1638
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.00047,
+      "loss": 0.6932,
+      "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,
+      "step": 2457
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.00046,
+      "loss": 0.6932,
+      "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,
+      "step": 3276
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 0.00045000000000000004,
+      "loss": 0.6932,
+      "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,
+      "step": 4095
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.00044,
+      "loss": 0.6932,
+      "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,
+      "step": 4914
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 0.00043,
+      "loss": 0.6932,
+      "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,
+      "step": 5733
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.00042,
+      "loss": 0.6931,
+      "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,
+      "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
+    }
+  ],
+  "max_steps": 40950,
+  "num_train_epochs": 50,
+  "total_flos": 1.2409680077573325e+17,
+  "trial_name": null,
+  "trial_params": null
+}