End of training

README.md

@@ -1,14 +1,29 @@
 ---
 library_name: transformers
+language:
+- en
 license: apache-2.0
 base_model: google/bert_uncased_L-4_H-128_A-2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: bert_uncased_L-4_H-128_A-2_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.7331163547599675
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_uncased_L-4_H-128_A-2_mnli
 
-This model is a fine-tuned version of [google/bert_uncased_L-4_H-128_A-2](https://huggingface.co/google/bert_uncased_L-4_H-128_A-2) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-4_H-128_A-2](https://huggingface.co/google/bert_uncased_L-4_H-128_A-2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6699
-- Accuracy: 0.7343
+- Loss: 0.6425
+- Accuracy: 0.7331
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 13.0,
+    "epoch_mm": 13.0,
+    "eval_accuracy": 0.7314314824248599,
+    "eval_accuracy_mm": 0.7331163547599675,
+    "eval_loss": 0.6530612111091614,
+    "eval_loss_mm": 0.6425328254699707,
+    "eval_runtime": 3.2268,
+    "eval_runtime_mm": 3.2612,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3041.719,
+    "eval_samples_per_second_mm": 3014.886,
+    "eval_steps_per_second": 12.086,
+    "eval_steps_per_second_mm": 11.959,
+    "total_flos": 6353499042229248.0,
+    "train_loss": 0.6473511495103378,
+    "train_runtime": 2152.219,
+    "train_samples": 392702,
+    "train_samples_per_second": 9123.188,
+    "train_steps_per_second": 35.638
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 13.0,
+    "epoch_mm": 13.0,
+    "eval_accuracy": 0.7314314824248599,
+    "eval_accuracy_mm": 0.7331163547599675,
+    "eval_loss": 0.6530612111091614,
+    "eval_loss_mm": 0.6425328254699707,
+    "eval_runtime": 3.2268,
+    "eval_runtime_mm": 3.2612,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3041.719,
+    "eval_samples_per_second_mm": 3014.886,
+    "eval_steps_per_second": 12.086,
+    "eval_steps_per_second_mm": 11.959
+}

logs/events.out.tfevents.1733341143.ki-g0008.1764072.35

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a04e6f9bd3640f34548a12ce5c5a5959b03d51b2e9fc8867d1794d28d43c005e
+size 746

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 13.0,
+    "total_flos": 6353499042229248.0,
+    "train_loss": 0.6473511495103378,
+    "train_runtime": 2152.219,
+    "train_samples": 392702,
+    "train_samples_per_second": 9123.188,
+    "train_steps_per_second": 35.638
+}

trainer_state.json

@@ -0,0 +1,259 @@
+{
+  "best_metric": 0.6530612111091614,
+  "best_model_checkpoint": "bert_uncased_L-4_H-128_A-2_mnli/checkpoint-12272",
+  "epoch": 13.0,
+  "eval_steps": 500,
+  "global_step": 19942,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.6993112564086914,
+      "learning_rate": 4.9e-05,
+      "loss": 0.8752,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6558329088130412,
+      "eval_loss": 0.7743282318115234,
+      "eval_runtime": 3.4721,
+      "eval_samples_per_second": 2826.817,
+      "eval_steps_per_second": 11.232,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 2.562037944793701,
+      "learning_rate": 4.8e-05,
+      "loss": 0.7714,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6856851757514009,
+      "eval_loss": 0.7263490557670593,
+      "eval_runtime": 3.284,
+      "eval_samples_per_second": 2988.699,
+      "eval_steps_per_second": 11.876,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 2.291146993637085,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7255,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7019867549668874,
+      "eval_loss": 0.6945955157279968,
+      "eval_runtime": 3.276,
+      "eval_samples_per_second": 2996.017,
+      "eval_steps_per_second": 11.905,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 2.2317264080047607,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.6927,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7087111563932756,
+      "eval_loss": 0.6788806915283203,
+      "eval_runtime": 3.3205,
+      "eval_samples_per_second": 2955.889,
+      "eval_steps_per_second": 11.745,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 2.6884801387786865,
+      "learning_rate": 4.5e-05,
+      "loss": 0.6662,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7205298013245033,
+      "eval_loss": 0.6656607985496521,
+      "eval_runtime": 3.3278,
+      "eval_samples_per_second": 2949.42,
+      "eval_steps_per_second": 11.72,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.7260897159576416,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6441,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7228731533367295,
+      "eval_loss": 0.6691338419914246,
+      "eval_runtime": 3.2556,
+      "eval_samples_per_second": 3014.76,
+      "eval_steps_per_second": 11.979,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.3097150325775146,
+      "learning_rate": 4.3e-05,
+      "loss": 0.625,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7258278145695364,
+      "eval_loss": 0.6622374057769775,
+      "eval_runtime": 3.2835,
+      "eval_samples_per_second": 2989.191,
+      "eval_steps_per_second": 11.878,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.3115994930267334,
+      "learning_rate": 4.2e-05,
+      "loss": 0.607,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7314314824248599,
+      "eval_loss": 0.6530612111091614,
+      "eval_runtime": 3.1955,
+      "eval_samples_per_second": 3071.492,
+      "eval_steps_per_second": 12.205,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.484405994415283,
+      "learning_rate": 4.1e-05,
+      "loss": 0.5894,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.7308201732042792,
+      "eval_loss": 0.6612954139709473,
+      "eval_runtime": 3.2486,
+      "eval_samples_per_second": 3021.261,
+      "eval_steps_per_second": 12.005,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 2.5099828243255615,
+      "learning_rate": 4e-05,
+      "loss": 0.5754,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.7292919001528273,
+      "eval_loss": 0.6590717434883118,
+      "eval_runtime": 3.2884,
+      "eval_samples_per_second": 2984.725,
+      "eval_steps_per_second": 11.86,
+      "step": 15340
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 2.7104341983795166,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.5615,
+      "step": 16874
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.7286805909322466,
+      "eval_loss": 0.6634514927864075,
+      "eval_runtime": 3.2403,
+      "eval_samples_per_second": 3029.047,
+      "eval_steps_per_second": 12.036,
+      "step": 16874
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 2.729738473892212,
+      "learning_rate": 3.8e-05,
+      "loss": 0.5477,
+      "step": 18408
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.7343861436576669,
+      "eval_loss": 0.6701191663742065,
+      "eval_runtime": 3.2255,
+      "eval_samples_per_second": 3042.947,
+      "eval_steps_per_second": 12.091,
+      "step": 18408
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 3.311851978302002,
+      "learning_rate": 3.7e-05,
+      "loss": 0.5343,
+      "step": 19942
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.7342842587875701,
+      "eval_loss": 0.6698570847511292,
+      "eval_runtime": 3.2844,
+      "eval_samples_per_second": 2988.397,
+      "eval_steps_per_second": 11.874,
+      "step": 19942
+    },
+    {
+      "epoch": 13.0,
+      "step": 19942,
+      "total_flos": 6353499042229248.0,
+      "train_loss": 0.6473511495103378,
+      "train_runtime": 2152.219,
+      "train_samples_per_second": 9123.188,
+      "train_steps_per_second": 35.638
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 76700,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 5,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 5
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6353499042229248.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}