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-512_A-8
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - spearmanr
 model-index:
 - name: bert_uncased_L-4_H-512_A-8_stsb
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE STSB
+      type: glue
+      args: stsb
+    metrics:
+    - name: Spearmanr
+      type: spearmanr
+      value: 0.8704339623572931
 ---
 
 <!-- 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-512_A-8_stsb
 
-This model is a fine-tuned version of [google/bert_uncased_L-4_H-512_A-8](https://huggingface.co/google/bert_uncased_L-4_H-512_A-8) on an unknown dataset.
+This model is a fine-tuned version of [google/bert_uncased_L-4_H-512_A-8](https://huggingface.co/google/bert_uncased_L-4_H-512_A-8) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5781
-- Pearson: 0.8748
-- Spearmanr: 0.8713
-- Combined Score: 0.8731
+- Loss: 0.5646
+- Pearson: 0.8739
+- Spearmanr: 0.8704
+- Combined Score: 0.8721
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 23.0,
+    "eval_combined_score": 0.8721440455693346,
+    "eval_loss": 0.5645565390586853,
+    "eval_pearson": 0.873854128781376,
+    "eval_runtime": 0.6286,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 2386.301,
+    "eval_spearmanr": 0.8704339623572931,
+    "eval_steps_per_second": 9.545,
+    "total_flos": 2614663011045888.0,
+    "train_loss": 0.3415615337782862,
+    "train_runtime": 113.3247,
+    "train_samples": 5749,
+    "train_samples_per_second": 2536.516,
+    "train_steps_per_second": 10.148
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 23.0,
+    "eval_combined_score": 0.8721440455693346,
+    "eval_loss": 0.5645565390586853,
+    "eval_pearson": 0.873854128781376,
+    "eval_runtime": 0.6286,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 2386.301,
+    "eval_spearmanr": 0.8704339623572931,
+    "eval_steps_per_second": 9.545
+}

logs/events.out.tfevents.1733338323.ki-g0008.1761130.31

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1526038a2ec44c8e4d2d9e6fea6e74797aaed133746e2ef9ac7f0b95d6bb4d6
+size 521

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 23.0,
+    "total_flos": 2614663011045888.0,
+    "train_loss": 0.3415615337782862,
+    "train_runtime": 113.3247,
+    "train_samples": 5749,
+    "train_samples_per_second": 2536.516,
+    "train_steps_per_second": 10.148
+}

trainer_state.json

@@ -0,0 +1,465 @@
+{
+  "best_metric": 0.5645565390586853,
+  "best_model_checkpoint": "bert_uncased_L-4_H-512_A-8_stsb/checkpoint-414",
+  "epoch": 23.0,
+  "eval_steps": 500,
+  "global_step": 529,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 6.830661773681641,
+      "learning_rate": 4.9e-05,
+      "loss": 2.5878,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.7701561984669311,
+      "eval_loss": 0.9754181504249573,
+      "eval_pearson": 0.7823353387435045,
+      "eval_runtime": 0.6067,
+      "eval_samples_per_second": 2472.44,
+      "eval_spearmanr": 0.7579770581903577,
+      "eval_steps_per_second": 9.89,
+      "step": 23
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 5.863245487213135,
+      "learning_rate": 4.8e-05,
+      "loss": 0.797,
+      "step": 46
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.8474045759571504,
+      "eval_loss": 0.7765601277351379,
+      "eval_pearson": 0.8466114574812236,
+      "eval_runtime": 0.5993,
+      "eval_samples_per_second": 2502.798,
+      "eval_spearmanr": 0.8481976944330771,
+      "eval_steps_per_second": 10.011,
+      "step": 46
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 5.94612979888916,
+      "learning_rate": 4.7e-05,
+      "loss": 0.5786,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.8595259972012628,
+      "eval_loss": 0.6313804984092712,
+      "eval_pearson": 0.8603476034751536,
+      "eval_runtime": 0.5837,
+      "eval_samples_per_second": 2569.683,
+      "eval_spearmanr": 0.8587043909273718,
+      "eval_steps_per_second": 10.279,
+      "step": 69
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 5.815054416656494,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.4961,
+      "step": 92
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.863972207025754,
+      "eval_loss": 0.6342185139656067,
+      "eval_pearson": 0.864253629360168,
+      "eval_runtime": 0.5999,
+      "eval_samples_per_second": 2500.544,
+      "eval_spearmanr": 0.8636907846913399,
+      "eval_steps_per_second": 10.002,
+      "step": 92
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 5.26311731338501,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3944,
+      "step": 115
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.8688625775742702,
+      "eval_loss": 0.6018186807632446,
+      "eval_pearson": 0.8694015311403887,
+      "eval_runtime": 0.5948,
+      "eval_samples_per_second": 2521.799,
+      "eval_spearmanr": 0.8683236240081519,
+      "eval_steps_per_second": 10.087,
+      "step": 115
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 5.787060737609863,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3362,
+      "step": 138
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.8657998136907656,
+      "eval_loss": 0.6101198792457581,
+      "eval_pearson": 0.8658762558868609,
+      "eval_runtime": 0.596,
+      "eval_samples_per_second": 2516.574,
+      "eval_spearmanr": 0.8657233714946704,
+      "eval_steps_per_second": 10.066,
+      "step": 138
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 11.731350898742676,
+      "learning_rate": 4.3e-05,
+      "loss": 0.2932,
+      "step": 161
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.867202263600297,
+      "eval_loss": 0.6055566072463989,
+      "eval_pearson": 0.8677929726947374,
+      "eval_runtime": 0.6269,
+      "eval_samples_per_second": 2392.606,
+      "eval_spearmanr": 0.8666115545058564,
+      "eval_steps_per_second": 9.57,
+      "step": 161
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 6.97740364074707,
+      "learning_rate": 4.2e-05,
+      "loss": 0.2495,
+      "step": 184
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.8675369208706127,
+      "eval_loss": 0.6255138516426086,
+      "eval_pearson": 0.8678571378034932,
+      "eval_runtime": 0.6161,
+      "eval_samples_per_second": 2434.767,
+      "eval_spearmanr": 0.8672167039377323,
+      "eval_steps_per_second": 9.739,
+      "step": 184
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 4.306568145751953,
+      "learning_rate": 4.1e-05,
+      "loss": 0.2268,
+      "step": 207
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.8691801970155717,
+      "eval_loss": 0.5970388054847717,
+      "eval_pearson": 0.8699055489590755,
+      "eval_runtime": 0.6258,
+      "eval_samples_per_second": 2396.782,
+      "eval_spearmanr": 0.8684548450720679,
+      "eval_steps_per_second": 9.587,
+      "step": 207
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 8.294599533081055,
+      "learning_rate": 4e-05,
+      "loss": 0.2037,
+      "step": 230
+    },
+    {
+      "epoch": 10.0,
+      "eval_combined_score": 0.868154308441244,
+      "eval_loss": 0.6516512632369995,
+      "eval_pearson": 0.869129623680379,
+      "eval_runtime": 0.6109,
+      "eval_samples_per_second": 2455.261,
+      "eval_spearmanr": 0.8671789932021089,
+      "eval_steps_per_second": 9.821,
+      "step": 230
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 4.237338542938232,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.191,
+      "step": 253
+    },
+    {
+      "epoch": 11.0,
+      "eval_combined_score": 0.8693045282961145,
+      "eval_loss": 0.601725697517395,
+      "eval_pearson": 0.8709460268802508,
+      "eval_runtime": 0.6389,
+      "eval_samples_per_second": 2347.804,
+      "eval_spearmanr": 0.8676630297119781,
+      "eval_steps_per_second": 9.391,
+      "step": 253
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 3.5681674480438232,
+      "learning_rate": 3.8e-05,
+      "loss": 0.1678,
+      "step": 276
+    },
+    {
+      "epoch": 12.0,
+      "eval_combined_score": 0.8694683064851385,
+      "eval_loss": 0.6096729636192322,
+      "eval_pearson": 0.8704163838060811,
+      "eval_runtime": 0.61,
+      "eval_samples_per_second": 2458.83,
+      "eval_spearmanr": 0.8685202291641958,
+      "eval_steps_per_second": 9.835,
+      "step": 276
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 3.010383367538452,
+      "learning_rate": 3.7e-05,
+      "loss": 0.1546,
+      "step": 299
+    },
+    {
+      "epoch": 13.0,
+      "eval_combined_score": 0.8707355017307861,
+      "eval_loss": 0.6051801443099976,
+      "eval_pearson": 0.8713328676887442,
+      "eval_runtime": 0.6283,
+      "eval_samples_per_second": 2387.444,
+      "eval_spearmanr": 0.8701381357728278,
+      "eval_steps_per_second": 9.55,
+      "step": 299
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 4.628120422363281,
+      "learning_rate": 3.6e-05,
+      "loss": 0.1486,
+      "step": 322
+    },
+    {
+      "epoch": 14.0,
+      "eval_combined_score": 0.8701183473370655,
+      "eval_loss": 0.5913792252540588,
+      "eval_pearson": 0.8713640146825573,
+      "eval_runtime": 0.6333,
+      "eval_samples_per_second": 2368.584,
+      "eval_spearmanr": 0.8688726799915738,
+      "eval_steps_per_second": 9.474,
+      "step": 322
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 3.274272918701172,
+      "learning_rate": 3.5e-05,
+      "loss": 0.1372,
+      "step": 345
+    },
+    {
+      "epoch": 15.0,
+      "eval_combined_score": 0.8719789615058817,
+      "eval_loss": 0.6174795031547546,
+      "eval_pearson": 0.8737840000829027,
+      "eval_runtime": 0.6364,
+      "eval_samples_per_second": 2357.008,
+      "eval_spearmanr": 0.8701739229288608,
+      "eval_steps_per_second": 9.428,
+      "step": 345
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 2.2427499294281006,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.131,
+      "step": 368
+    },
+    {
+      "epoch": 16.0,
+      "eval_combined_score": 0.8714552917935461,
+      "eval_loss": 0.5825785994529724,
+      "eval_pearson": 0.8726691216431404,
+      "eval_runtime": 0.6147,
+      "eval_samples_per_second": 2440.092,
+      "eval_spearmanr": 0.8702414619439519,
+      "eval_steps_per_second": 9.76,
+      "step": 368
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 3.5106945037841797,
+      "learning_rate": 3.3e-05,
+      "loss": 0.1216,
+      "step": 391
+    },
+    {
+      "epoch": 17.0,
+      "eval_combined_score": 0.8701714131214551,
+      "eval_loss": 0.577867865562439,
+      "eval_pearson": 0.8717439862070048,
+      "eval_runtime": 0.5982,
+      "eval_samples_per_second": 2507.56,
+      "eval_spearmanr": 0.8685988400359054,
+      "eval_steps_per_second": 10.03,
+      "step": 391
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 4.543298721313477,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.1145,
+      "step": 414
+    },
+    {
+      "epoch": 18.0,
+      "eval_combined_score": 0.8721440455693346,
+      "eval_loss": 0.5645565390586853,
+      "eval_pearson": 0.873854128781376,
+      "eval_runtime": 0.6076,
+      "eval_samples_per_second": 2468.855,
+      "eval_spearmanr": 0.8704339623572931,
+      "eval_steps_per_second": 9.875,
+      "step": 414
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 4.314229965209961,
+      "learning_rate": 3.1e-05,
+      "loss": 0.1158,
+      "step": 437
+    },
+    {
+      "epoch": 19.0,
+      "eval_combined_score": 0.8724136267007434,
+      "eval_loss": 0.5810861587524414,
+      "eval_pearson": 0.8737710324523099,
+      "eval_runtime": 0.6002,
+      "eval_samples_per_second": 2499.237,
+      "eval_spearmanr": 0.871056220949177,
+      "eval_steps_per_second": 9.997,
+      "step": 437
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 5.715078830718994,
+      "learning_rate": 3e-05,
+      "loss": 0.109,
+      "step": 460
+    },
+    {
+      "epoch": 20.0,
+      "eval_combined_score": 0.8741776682213974,
+      "eval_loss": 0.5896406173706055,
+      "eval_pearson": 0.876321941342814,
+      "eval_runtime": 0.599,
+      "eval_samples_per_second": 2504.307,
+      "eval_spearmanr": 0.8720333950999808,
+      "eval_steps_per_second": 10.017,
+      "step": 460
+    },
+    {
+      "epoch": 21.0,
+      "grad_norm": 3.1881017684936523,
+      "learning_rate": 2.9e-05,
+      "loss": 0.105,
+      "step": 483
+    },
+    {
+      "epoch": 21.0,
+      "eval_combined_score": 0.8720974137486168,
+      "eval_loss": 0.5863229036331177,
+      "eval_pearson": 0.8737022748828657,
+      "eval_runtime": 0.5928,
+      "eval_samples_per_second": 2530.409,
+      "eval_spearmanr": 0.8704925526143681,
+      "eval_steps_per_second": 10.122,
+      "step": 483
+    },
+    {
+      "epoch": 22.0,
+      "grad_norm": 3.1537728309631348,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 0.0995,
+      "step": 506
+    },
+    {
+      "epoch": 22.0,
+      "eval_combined_score": 0.8721144188715132,
+      "eval_loss": 0.5757761001586914,
+      "eval_pearson": 0.8740875105899543,
+      "eval_runtime": 0.6179,
+      "eval_samples_per_second": 2427.472,
+      "eval_spearmanr": 0.8701413271530721,
+      "eval_steps_per_second": 9.71,
+      "step": 506
+    },
+    {
+      "epoch": 23.0,
+      "grad_norm": 2.4976413249969482,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 0.0971,
+      "step": 529
+    },
+    {
+      "epoch": 23.0,
+      "eval_combined_score": 0.8730602670988505,
+      "eval_loss": 0.5780627727508545,
+      "eval_pearson": 0.8748200461236054,
+      "eval_runtime": 0.613,
+      "eval_samples_per_second": 2446.918,
+      "eval_spearmanr": 0.8713004880740955,
+      "eval_steps_per_second": 9.788,
+      "step": 529
+    },
+    {
+      "epoch": 23.0,
+      "step": 529,
+      "total_flos": 2614663011045888.0,
+      "train_loss": 0.3415615337782862,
+      "train_runtime": 113.3247,
+      "train_samples_per_second": 2536.516,
+      "train_steps_per_second": 10.148
+    }
+  ],
+  "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": 2614663011045888.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}