End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_5_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - spearmanr
 model-index:
 - name: tiny_bert_km_5_v2_stsb
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE STSB
+      type: glue
+      args: stsb
+    metrics:
+    - name: Spearmanr
+      type: spearmanr
+      value: 0.2086240433941912
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,12 +30,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_km_5_v2_stsb
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_5_v2](https://huggingface.co/Hartunka/tiny_bert_km_5_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_5_v2](https://huggingface.co/Hartunka/tiny_bert_km_5_v2) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.5246
-- Pearson: 0.2912
-- Spearmanr: 0.2977
-- Combined Score: 0.2944
+- Loss: 2.3479
+- Pearson: 0.2208
+- Spearmanr: 0.2086
+- Combined Score: 0.2147
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 9.0,
+    "eval_combined_score": 0.21473410605933732,
+    "eval_loss": 2.3479087352752686,
+    "eval_pearson": 0.22084416872448345,
+    "eval_runtime": 0.4695,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 3194.956,
+    "eval_spearmanr": 0.2086240433941912,
+    "eval_steps_per_second": 12.78,
+    "total_flos": 1356793384840704.0,
+    "train_loss": 1.6471541676544337,
+    "train_runtime": 37.5069,
+    "train_samples": 5749,
+    "train_samples_per_second": 7663.93,
+    "train_steps_per_second": 30.661
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 9.0,
+    "eval_combined_score": 0.21473410605933732,
+    "eval_loss": 2.3479087352752686,
+    "eval_pearson": 0.22084416872448345,
+    "eval_runtime": 0.4695,
+    "eval_samples": 1500,
+    "eval_samples_per_second": 3194.956,
+    "eval_spearmanr": 0.2086240433941912,
+    "eval_steps_per_second": 12.78
+}

logs/events.out.tfevents.1745270808.s_005_m.2836625.13

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d6bbe716168385c5e34d302336c0316b3569a5b5c68cd5841579791cdadd491
+size 473

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.0,
+    "total_flos": 1356793384840704.0,
+    "train_loss": 1.6471541676544337,
+    "train_runtime": 37.5069,
+    "train_samples": 5749,
+    "train_samples_per_second": 7663.93,
+    "train_steps_per_second": 30.661
+}

trainer_state.json

@@ -0,0 +1,214 @@
+{
+  "best_global_step": 92,
+  "best_metric": 2.3479087352752686,
+  "best_model_checkpoint": "tiny_bert_km_5_v2_stsb/checkpoint-92",
+  "epoch": 9.0,
+  "eval_steps": 500,
+  "global_step": 207,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 3.2017595767974854,
+      "learning_rate": 4.9e-05,
+      "loss": 2.7925,
+      "step": 23
+    },
+    {
+      "epoch": 1.0,
+      "eval_combined_score": 0.11189326293514337,
+      "eval_loss": 2.3609859943389893,
+      "eval_pearson": 0.11928981153434301,
+      "eval_runtime": 0.4356,
+      "eval_samples_per_second": 3443.807,
+      "eval_spearmanr": 0.10449671433594372,
+      "eval_steps_per_second": 13.775,
+      "step": 23
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.0342791080474854,
+      "learning_rate": 4.8e-05,
+      "loss": 2.0233,
+      "step": 46
+    },
+    {
+      "epoch": 2.0,
+      "eval_combined_score": 0.10503262884435496,
+      "eval_loss": 2.7264516353607178,
+      "eval_pearson": 0.10536681667826245,
+      "eval_runtime": 0.4819,
+      "eval_samples_per_second": 3112.724,
+      "eval_spearmanr": 0.10469844101044747,
+      "eval_steps_per_second": 12.451,
+      "step": 46
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.3021488189697266,
+      "learning_rate": 4.7e-05,
+      "loss": 1.9375,
+      "step": 69
+    },
+    {
+      "epoch": 3.0,
+      "eval_combined_score": 0.15795839329801392,
+      "eval_loss": 2.3723838329315186,
+      "eval_pearson": 0.16413878451371774,
+      "eval_runtime": 0.5006,
+      "eval_samples_per_second": 2996.464,
+      "eval_spearmanr": 0.15177800208231013,
+      "eval_steps_per_second": 11.986,
+      "step": 69
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 12.65180778503418,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 1.8081,
+      "step": 92
+    },
+    {
+      "epoch": 4.0,
+      "eval_combined_score": 0.21473410605933732,
+      "eval_loss": 2.3479087352752686,
+      "eval_pearson": 0.22084416872448345,
+      "eval_runtime": 0.4472,
+      "eval_samples_per_second": 3353.842,
+      "eval_spearmanr": 0.2086240433941912,
+      "eval_steps_per_second": 13.415,
+      "step": 92
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 14.563904762268066,
+      "learning_rate": 4.5e-05,
+      "loss": 1.593,
+      "step": 115
+    },
+    {
+      "epoch": 5.0,
+      "eval_combined_score": 0.2341727272115602,
+      "eval_loss": 2.5032310485839844,
+      "eval_pearson": 0.23463414846011482,
+      "eval_runtime": 0.4639,
+      "eval_samples_per_second": 3233.596,
+      "eval_spearmanr": 0.2337113059630056,
+      "eval_steps_per_second": 12.934,
+      "step": 115
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 16.085830688476562,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 1.4271,
+      "step": 138
+    },
+    {
+      "epoch": 6.0,
+      "eval_combined_score": 0.2668894994914338,
+      "eval_loss": 2.3829503059387207,
+      "eval_pearson": 0.2663211377699541,
+      "eval_runtime": 0.461,
+      "eval_samples_per_second": 3253.99,
+      "eval_spearmanr": 0.26745786121291354,
+      "eval_steps_per_second": 13.016,
+      "step": 138
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 17.25328826904297,
+      "learning_rate": 4.3e-05,
+      "loss": 1.2196,
+      "step": 161
+    },
+    {
+      "epoch": 7.0,
+      "eval_combined_score": 0.3036982340745887,
+      "eval_loss": 2.367271900177002,
+      "eval_pearson": 0.3016266607369947,
+      "eval_runtime": 0.4595,
+      "eval_samples_per_second": 3264.444,
+      "eval_spearmanr": 0.3057698074121827,
+      "eval_steps_per_second": 13.058,
+      "step": 161
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 19.59164810180664,
+      "learning_rate": 4.2e-05,
+      "loss": 1.0511,
+      "step": 184
+    },
+    {
+      "epoch": 8.0,
+      "eval_combined_score": 0.29420108006699475,
+      "eval_loss": 2.502896785736084,
+      "eval_pearson": 0.2914824122831241,
+      "eval_runtime": 0.5124,
+      "eval_samples_per_second": 2927.296,
+      "eval_spearmanr": 0.2969197478508654,
+      "eval_steps_per_second": 11.709,
+      "step": 184
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 13.128865242004395,
+      "learning_rate": 4.1e-05,
+      "loss": 0.9721,
+      "step": 207
+    },
+    {
+      "epoch": 9.0,
+      "eval_combined_score": 0.29443550792517503,
+      "eval_loss": 2.524557113647461,
+      "eval_pearson": 0.2911705345450899,
+      "eval_runtime": 0.4558,
+      "eval_samples_per_second": 3290.86,
+      "eval_spearmanr": 0.29770048130526017,
+      "eval_steps_per_second": 13.163,
+      "step": 207
+    },
+    {
+      "epoch": 9.0,
+      "step": 207,
+      "total_flos": 1356793384840704.0,
+      "train_loss": 1.6471541676544337,
+      "train_runtime": 37.5069,
+      "train_samples_per_second": 7663.93,
+      "train_steps_per_second": 30.661
+    }
+  ],
+  "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": 1356793384840704.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}