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-256_A-4
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - spearmanr
 model-index:
 - name: bert_uncased_L-4_H-256_A-4_stsb
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE STSB
+      type: glue
+      args: stsb
+    metrics:
+    - name: Spearmanr
+      type: spearmanr
+      value: 0.8541619713648296
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -16,12 +31,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_uncased_L-4_H-256_A-4_stsb
 
-This model is a fine-tuned version of [google/bert_uncased_L-4_H-256_A-4](https://huggingface.co/google/bert_uncased_L-4_H-256_A-4) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-4_H-256_A-4](https://huggingface.co/google/bert_uncased_L-4_H-256_A-4) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6379
-- Pearson: 0.8559
-- Spearmanr: 0.8555
-- Combined Score: 0.8557
+- Loss: 0.6283
+- Pearson: 0.8545
+- Spearmanr: 0.8542
+- Combined Score: 0.8543
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 16.0,
+    "eval_combined_score": 0.8543458921269349,
+    "eval_loss": 0.628333568572998,
+    "eval_pearson": 0.8545298128890402,
+    "eval_runtime": 0.5349,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 2804.436,
+    "eval_spearmanr": 0.8541619713648296,
+    "eval_steps_per_second": 11.218,
+    "total_flos": 455736856141824.0,
+    "train_loss": 0.9891756526801897,
+    "train_runtime": 56.9981,
+    "train_samples": 5749,
+    "train_samples_per_second": 5043.151,
+    "train_steps_per_second": 20.176
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 16.0,
+    "eval_combined_score": 0.8543458921269349,
+    "eval_loss": 0.628333568572998,
+    "eval_pearson": 0.8545298128890402,
+    "eval_runtime": 0.5349,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 2804.436,
+    "eval_spearmanr": 0.8541619713648296,
+    "eval_steps_per_second": 11.218
+}

logs/events.out.tfevents.1733333039.ki-g0008.1761130.13

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4f09888fc80470538a05a3a89917ed230adbe79ae4c15e4343fb4f439af634f
+size 521

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 16.0,
+    "total_flos": 455736856141824.0,
+    "train_loss": 0.9891756526801897,
+    "train_runtime": 56.9981,
+    "train_samples": 5749,
+    "train_samples_per_second": 5043.151,
+    "train_steps_per_second": 20.176
+}

trainer_state.json

@@ -0,0 +1,339 @@
+{
+  "best_metric": 0.628333568572998,
+  "best_model_checkpoint": "bert_uncased_L-4_H-256_A-4_stsb/checkpoint-253",
+  "epoch": 16.0,
+  "eval_steps": 500,
+  "global_step": 368,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 23.512914657592773,
+      "learning_rate": 4.9e-05,
+      "loss": 5.5773,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.35940682306384497,
+      "eval_loss": 2.741238594055176,
+      "eval_pearson": 0.3844718888057109,
+      "eval_runtime": 0.5362,
+      "eval_samples_per_second": 2797.44,
+      "eval_spearmanr": 0.334341757321979,
+      "eval_steps_per_second": 11.19,
+      "step": 23
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 15.94455623626709,
+      "learning_rate": 4.8e-05,
+      "loss": 2.5793,
+      "step": 46
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.7641849595072145,
+      "eval_loss": 1.9157546758651733,
+      "eval_pearson": 0.7726894274450853,
+      "eval_runtime": 0.5382,
+      "eval_samples_per_second": 2787.176,
+      "eval_spearmanr": 0.7556804915693437,
+      "eval_steps_per_second": 11.149,
+      "step": 46
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 10.711121559143066,
+      "learning_rate": 4.7e-05,
+      "loss": 1.5767,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.7589673053156418,
+      "eval_loss": 0.9540784358978271,
+      "eval_pearson": 0.7705953786726122,
+      "eval_runtime": 0.537,
+      "eval_samples_per_second": 2793.104,
+      "eval_spearmanr": 0.7473392319586716,
+      "eval_steps_per_second": 11.172,
+      "step": 69
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 7.894664764404297,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.9474,
+      "step": 92
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.8101368624163152,
+      "eval_loss": 0.7628255486488342,
+      "eval_pearson": 0.8132575634904194,
+      "eval_runtime": 0.538,
+      "eval_samples_per_second": 2787.885,
+      "eval_spearmanr": 0.8070161613422111,
+      "eval_steps_per_second": 11.152,
+      "step": 92
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 7.180455684661865,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7258,
+      "step": 115
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.8405981737206667,
+      "eval_loss": 0.6784510016441345,
+      "eval_pearson": 0.8382918173715312,
+      "eval_runtime": 0.5415,
+      "eval_samples_per_second": 2770.208,
+      "eval_spearmanr": 0.8429045300698023,
+      "eval_steps_per_second": 11.081,
+      "step": 115
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 8.510337829589844,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6162,
+      "step": 138
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.8437314098589771,
+      "eval_loss": 0.6755565404891968,
+      "eval_pearson": 0.8435713950894774,
+      "eval_runtime": 0.5333,
+      "eval_samples_per_second": 2812.417,
+      "eval_spearmanr": 0.8438914246284768,
+      "eval_steps_per_second": 11.25,
+      "step": 138
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.887516498565674,
+      "learning_rate": 4.3e-05,
+      "loss": 0.5455,
+      "step": 161
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.849162211616322,
+      "eval_loss": 0.6391196250915527,
+      "eval_pearson": 0.8479729963672471,
+      "eval_runtime": 0.5344,
+      "eval_samples_per_second": 2806.816,
+      "eval_spearmanr": 0.850351426865397,
+      "eval_steps_per_second": 11.227,
+      "step": 161
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 8.618912696838379,
+      "learning_rate": 4.2e-05,
+      "loss": 0.4912,
+      "step": 184
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.8466254837460676,
+      "eval_loss": 0.6581634879112244,
+      "eval_pearson": 0.8460862349631103,
+      "eval_runtime": 0.539,
+      "eval_samples_per_second": 2783.12,
+      "eval_spearmanr": 0.8471647325290248,
+      "eval_steps_per_second": 11.132,
+      "step": 184
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 6.421316146850586,
+      "learning_rate": 4.1e-05,
+      "loss": 0.4443,
+      "step": 207
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.8476912767928249,
+      "eval_loss": 0.6560913324356079,
+      "eval_pearson": 0.8471629476366933,
+      "eval_runtime": 0.5436,
+      "eval_samples_per_second": 2759.539,
+      "eval_spearmanr": 0.8482196059489565,
+      "eval_steps_per_second": 11.038,
+      "step": 207
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 6.827481269836426,
+      "learning_rate": 4e-05,
+      "loss": 0.3995,
+      "step": 230
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.8503318094261108,
+      "eval_loss": 0.6429010629653931,
+      "eval_pearson": 0.850358057284851,
+      "eval_runtime": 0.5319,
+      "eval_samples_per_second": 2820.144,
+      "eval_spearmanr": 0.8503055615673704,
+      "eval_steps_per_second": 11.281,
+      "step": 230
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 6.088059902191162,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.3689,
+      "step": 253
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.8543458921269349,
+      "eval_loss": 0.628333568572998,
+      "eval_pearson": 0.8545298128890402,
+      "eval_runtime": 0.5445,
+      "eval_samples_per_second": 2754.594,
+      "eval_spearmanr": 0.8541619713648296,
+      "eval_steps_per_second": 11.018,
+      "step": 253
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 4.50377893447876,
+      "learning_rate": 3.8e-05,
+      "loss": 0.3418,
+      "step": 276
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.8520085335126549,
+      "eval_loss": 0.6592078804969788,
+      "eval_pearson": 0.8519775414066034,
+      "eval_runtime": 0.5357,
+      "eval_samples_per_second": 2800.12,
+      "eval_spearmanr": 0.8520395256187064,
+      "eval_steps_per_second": 11.2,
+      "step": 276
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 4.240983486175537,
+      "learning_rate": 3.7e-05,
+      "loss": 0.3302,
+      "step": 299
+    },
+    {
+      "epoch": 13.0,
+      "eval_combined_score": 0.8527115311132238,
+      "eval_loss": 0.650736391544342,
+      "eval_pearson": 0.8524420075355381,
+      "eval_runtime": 0.5456,
+      "eval_samples_per_second": 2749.213,
+      "eval_spearmanr": 0.8529810546909096,
+      "eval_steps_per_second": 10.997,
+      "step": 299
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 7.435579776763916,
+      "learning_rate": 3.6e-05,
+      "loss": 0.319,
+      "step": 322
+    },
+    {
+      "epoch": 14.0,
+      "eval_combined_score": 0.8527060192476493,
+      "eval_loss": 0.648432195186615,
+      "eval_pearson": 0.8527872374234446,
+      "eval_runtime": 0.5301,
+      "eval_samples_per_second": 2829.767,
+      "eval_spearmanr": 0.852624801071854,
+      "eval_steps_per_second": 11.319,
+      "step": 322
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 5.8321027755737305,
+      "learning_rate": 3.5e-05,
+      "loss": 0.2863,
+      "step": 345
+    },
+    {
+      "epoch": 15.0,
+      "eval_combined_score": 0.8526379883202206,
+      "eval_loss": 0.639711320400238,
+      "eval_pearson": 0.8525586066146398,
+      "eval_runtime": 0.5532,
+      "eval_samples_per_second": 2711.331,
+      "eval_spearmanr": 0.8527173700258016,
+      "eval_steps_per_second": 10.845,
+      "step": 345
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 4.939488887786865,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.2774,
+      "step": 368
+    },
+    {
+      "epoch": 16.0,
+      "eval_combined_score": 0.855685351715838,
+      "eval_loss": 0.6379477977752686,
+      "eval_pearson": 0.8558522598630551,
+      "eval_runtime": 0.5326,
+      "eval_samples_per_second": 2816.114,
+      "eval_spearmanr": 0.855518443568621,
+      "eval_steps_per_second": 11.264,
+      "step": 368
+    },
+    {
+      "epoch": 16.0,
+      "step": 368,
+      "total_flos": 455736856141824.0,
+      "train_loss": 0.9891756526801897,
+      "train_runtime": 56.9981,
+      "train_samples_per_second": 5043.151,
+      "train_steps_per_second": 20.176
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1150,
+  "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": 455736856141824.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}