End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_100_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_km_100_v2_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6173718470301057
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_km_100_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_100_v2](https://huggingface.co/Hartunka/tiny_bert_km_100_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_100_v2](https://huggingface.co/Hartunka/tiny_bert_km_100_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9786
-- Accuracy: 0.6191
+- Loss: 0.8532
+- Accuracy: 0.6174
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6161996943453897,
+    "eval_accuracy_mm": 0.6173718470301057,
+    "eval_loss": 0.8644183278083801,
+    "eval_loss_mm": 0.853205144405365,
+    "eval_runtime": 2.8034,
+    "eval_runtime_mm": 2.8898,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3501.136,
+    "eval_samples_per_second_mm": 3402.285,
+    "eval_steps_per_second": 13.912,
+    "eval_steps_per_second_mm": 13.496,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.761610681168259,
+    "train_runtime": 2062.3602,
+    "train_samples": 392702,
+    "train_samples_per_second": 9520.694,
+    "train_steps_per_second": 37.19
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6161996943453897,
+    "eval_accuracy_mm": 0.6173718470301057,
+    "eval_loss": 0.8644183278083801,
+    "eval_loss_mm": 0.853205144405365,
+    "eval_runtime": 2.8034,
+    "eval_runtime_mm": 2.8898,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3501.136,
+    "eval_samples_per_second_mm": 3402.285,
+    "eval_steps_per_second": 13.912,
+    "eval_steps_per_second_mm": 13.496
+}

logs/events.out.tfevents.1745292837.s_005_m.2836625.89

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29b286056ffb5e8dc5988e57e4e4e31c1905cc1c731be0c719c8c3246c8ec7f9
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.761610681168259,
+    "train_runtime": 2062.3602,
+    "train_samples": 392702,
+    "train_samples_per_second": 9520.694,
+    "train_steps_per_second": 37.19
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8644183278083801,
+  "best_model_checkpoint": "tiny_bert_km_100_v2_mnli/checkpoint-7670",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 15340,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.9140164852142334,
+      "learning_rate": 4.9e-05,
+      "loss": 1.0063,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.537238920020377,
+      "eval_loss": 0.947323739528656,
+      "eval_runtime": 2.8342,
+      "eval_samples_per_second": 3463.068,
+      "eval_steps_per_second": 13.761,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3851509094238281,
+      "learning_rate": 4.8e-05,
+      "loss": 0.916,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5816607233825777,
+      "eval_loss": 0.8941206336021423,
+      "eval_runtime": 2.7849,
+      "eval_samples_per_second": 3524.408,
+      "eval_steps_per_second": 14.004,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.438253402709961,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8629,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6017320427916455,
+      "eval_loss": 0.8664500117301941,
+      "eval_runtime": 2.7822,
+      "eval_samples_per_second": 3527.723,
+      "eval_steps_per_second": 14.017,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.5567783117294312,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8183,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6045848191543556,
+      "eval_loss": 0.8703648447990417,
+      "eval_runtime": 2.8078,
+      "eval_samples_per_second": 3495.602,
+      "eval_steps_per_second": 13.89,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7820237874984741,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7746,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6161996943453897,
+      "eval_loss": 0.8644183278083801,
+      "eval_runtime": 2.8011,
+      "eval_samples_per_second": 3503.969,
+      "eval_steps_per_second": 13.923,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.103147268295288,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7324,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6173204279164544,
+      "eval_loss": 0.8818243145942688,
+      "eval_runtime": 2.7802,
+      "eval_samples_per_second": 3530.372,
+      "eval_steps_per_second": 14.028,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.9933719635009766,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6902,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.628222109016811,
+      "eval_loss": 0.8827255368232727,
+      "eval_runtime": 2.8175,
+      "eval_samples_per_second": 3483.629,
+      "eval_steps_per_second": 13.842,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.3629040718078613,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6478,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6206826286296485,
+      "eval_loss": 0.9343836307525635,
+      "eval_runtime": 2.8042,
+      "eval_samples_per_second": 3500.116,
+      "eval_steps_per_second": 13.908,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.6676883697509766,
+      "learning_rate": 4.1e-05,
+      "loss": 0.6049,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6225165562913907,
+      "eval_loss": 0.9615498781204224,
+      "eval_runtime": 2.8505,
+      "eval_samples_per_second": 3443.197,
+      "eval_steps_per_second": 13.682,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 2.945707082748413,
+      "learning_rate": 4e-05,
+      "loss": 0.5628,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6190524707080999,
+      "eval_loss": 0.9786139130592346,
+      "eval_runtime": 2.8978,
+      "eval_samples_per_second": 3387.031,
+      "eval_steps_per_second": 13.458,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.761610681168259,
+      "train_runtime": 2062.3602,
+      "train_samples_per_second": 9520.694,
+      "train_steps_per_second": 37.19
+    }
+  ],
+  "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": 1.029836100834816e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}