End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_rand_10_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_rand_10_v1_mnli
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MNLI
+      type: glue
+      args: mnli
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6491049633848658
 ---
 
 <!-- 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_rand_10_v1_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_rand_10_v1](https://huggingface.co/Hartunka/tiny_bert_rand_10_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_rand_10_v1](https://huggingface.co/Hartunka/tiny_bert_rand_10_v1) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9181
-- Accuracy: 0.6486
+- Loss: 0.7983
+- Accuracy: 0.6491
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6484971981660723,
+    "eval_accuracy_mm": 0.6491049633848658,
+    "eval_loss": 0.8026694059371948,
+    "eval_loss_mm": 0.7982809543609619,
+    "eval_runtime": 2.9309,
+    "eval_runtime_mm": 2.9555,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3348.802,
+    "eval_samples_per_second_mm": 3326.733,
+    "eval_steps_per_second": 13.306,
+    "eval_steps_per_second_mm": 13.196,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7320704254985944,
+    "train_runtime": 2102.0078,
+    "train_samples": 392702,
+    "train_samples_per_second": 9341.117,
+    "train_steps_per_second": 36.489
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6484971981660723,
+    "eval_accuracy_mm": 0.6491049633848658,
+    "eval_loss": 0.8026694059371948,
+    "eval_loss_mm": 0.7982809543609619,
+    "eval_runtime": 2.9309,
+    "eval_runtime_mm": 2.9555,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3348.802,
+    "eval_samples_per_second_mm": 3326.733,
+    "eval_steps_per_second": 13.306,
+    "eval_steps_per_second_mm": 13.196
+}

logs/events.out.tfevents.1744652119.s_005_m.2701049.89

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74411aaf04e179054e0a3fd64a2c2e1255afd18df8e1d6088b54a4c7e2b5d26b
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7320704254985944,
+    "train_runtime": 2102.0078,
+    "train_samples": 392702,
+    "train_samples_per_second": 9341.117,
+    "train_steps_per_second": 36.489
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8026694059371948,
+  "best_model_checkpoint": "tiny_bert_rand_10_v1_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": 1.1323920488357544,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9909,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5458991339786041,
+      "eval_loss": 0.9388576149940491,
+      "eval_runtime": 2.9578,
+      "eval_samples_per_second": 3318.302,
+      "eval_steps_per_second": 13.185,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.4309430122375488,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9076,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5857361181864493,
+      "eval_loss": 0.8834566473960876,
+      "eval_runtime": 2.8775,
+      "eval_samples_per_second": 3410.984,
+      "eval_steps_per_second": 13.554,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.7712280750274658,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8477,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6206826286296485,
+      "eval_loss": 0.84859299659729,
+      "eval_runtime": 2.8826,
+      "eval_samples_per_second": 3404.873,
+      "eval_steps_per_second": 13.529,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.7048949003219604,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.7883,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6344370860927152,
+      "eval_loss": 0.8233076333999634,
+      "eval_runtime": 2.8707,
+      "eval_samples_per_second": 3419.067,
+      "eval_steps_per_second": 13.586,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7251155376434326,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7357,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6484971981660723,
+      "eval_loss": 0.8026694059371948,
+      "eval_runtime": 2.8551,
+      "eval_samples_per_second": 3437.692,
+      "eval_steps_per_second": 13.66,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.5030930042266846,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.6888,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6493122771268467,
+      "eval_loss": 0.8328536748886108,
+      "eval_runtime": 2.8585,
+      "eval_samples_per_second": 3433.666,
+      "eval_steps_per_second": 13.644,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.1141350269317627,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6483,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6502292409577177,
+      "eval_loss": 0.8231375217437744,
+      "eval_runtime": 2.8796,
+      "eval_samples_per_second": 3408.438,
+      "eval_steps_per_second": 13.543,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.4901773929595947,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6075,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6527763627101375,
+      "eval_loss": 0.8672935366630554,
+      "eval_runtime": 2.8773,
+      "eval_samples_per_second": 3411.242,
+      "eval_steps_per_second": 13.555,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.7526743412017822,
+      "learning_rate": 4.1e-05,
+      "loss": 0.5713,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6456444218033622,
+      "eval_loss": 0.8915270566940308,
+      "eval_runtime": 2.8729,
+      "eval_samples_per_second": 3416.464,
+      "eval_steps_per_second": 13.575,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 2.8318562507629395,
+      "learning_rate": 4e-05,
+      "loss": 0.5346,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6485990830361691,
+      "eval_loss": 0.9180997610092163,
+      "eval_runtime": 2.8709,
+      "eval_samples_per_second": 3418.795,
+      "eval_steps_per_second": 13.585,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.7320704254985944,
+      "train_runtime": 2102.0078,
+      "train_samples_per_second": 9341.117,
+      "train_steps_per_second": 36.489
+    }
+  ],
+  "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
+}