End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_km_50_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_km_50_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: 0.5986574450772986
 ---
 
 <!-- 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_50_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_km_50_v2](https://huggingface.co/Hartunka/tiny_bert_km_50_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_km_50_v2](https://huggingface.co/Hartunka/tiny_bert_km_50_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0082
-- Accuracy: 0.6065
+- Loss: 0.8773
+- Accuracy: 0.5987
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6070300560366786,
+    "eval_accuracy_mm": 0.5986574450772986,
+    "eval_loss": 0.8703352808952332,
+    "eval_loss_mm": 0.8772774934768677,
+    "eval_runtime": 2.823,
+    "eval_runtime_mm": 3.0183,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3476.752,
+    "eval_samples_per_second_mm": 3257.459,
+    "eval_steps_per_second": 13.815,
+    "eval_steps_per_second_mm": 12.921,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7659492064952229,
+    "train_runtime": 2074.047,
+    "train_samples": 392702,
+    "train_samples_per_second": 9467.047,
+    "train_steps_per_second": 36.981
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6070300560366786,
+    "eval_accuracy_mm": 0.5986574450772986,
+    "eval_loss": 0.8703352808952332,
+    "eval_loss_mm": 0.8772774934768677,
+    "eval_runtime": 2.823,
+    "eval_runtime_mm": 3.0183,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3476.752,
+    "eval_samples_per_second_mm": 3257.459,
+    "eval_steps_per_second": 13.815,
+    "eval_steps_per_second_mm": 12.921
+}

logs/events.out.tfevents.1745288044.s_005_m.2836625.71

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:760ccb427d66c859320bede707571faf2f2c5905fdcabe8b9c414595ac824c1a
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7659492064952229,
+    "train_runtime": 2074.047,
+    "train_samples": 392702,
+    "train_samples_per_second": 9467.047,
+    "train_steps_per_second": 36.981
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8703352808952332,
+  "best_model_checkpoint": "tiny_bert_km_50_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": 1.0268884897232056,
+      "learning_rate": 4.9e-05,
+      "loss": 1.0026,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5339786041772797,
+      "eval_loss": 0.9471800327301025,
+      "eval_runtime": 2.9599,
+      "eval_samples_per_second": 3316.032,
+      "eval_steps_per_second": 13.176,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3657726049423218,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9203,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5731023942944473,
+      "eval_loss": 0.9026528000831604,
+      "eval_runtime": 2.8075,
+      "eval_samples_per_second": 3496.041,
+      "eval_steps_per_second": 13.892,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.3580602407455444,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8675,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5897096281202241,
+      "eval_loss": 0.8775414228439331,
+      "eval_runtime": 2.8127,
+      "eval_samples_per_second": 3489.517,
+      "eval_steps_per_second": 13.866,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.4758466482162476,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8211,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5985736118186449,
+      "eval_loss": 0.8704348802566528,
+      "eval_runtime": 2.781,
+      "eval_samples_per_second": 3529.336,
+      "eval_steps_per_second": 14.024,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.6115995645523071,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7773,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6070300560366786,
+      "eval_loss": 0.8703352808952332,
+      "eval_runtime": 2.9126,
+      "eval_samples_per_second": 3369.849,
+      "eval_steps_per_second": 13.39,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.9444777965545654,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7352,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6116148751910341,
+      "eval_loss": 0.8939315676689148,
+      "eval_runtime": 2.8219,
+      "eval_samples_per_second": 3478.123,
+      "eval_steps_per_second": 13.82,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.15396785736084,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6939,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6115129903209373,
+      "eval_loss": 0.9022112488746643,
+      "eval_runtime": 2.8175,
+      "eval_samples_per_second": 3483.619,
+      "eval_steps_per_second": 13.842,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.317307233810425,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6533,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6104941416199694,
+      "eval_loss": 0.9489761590957642,
+      "eval_runtime": 2.8028,
+      "eval_samples_per_second": 3501.863,
+      "eval_steps_per_second": 13.915,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 2.4730122089385986,
+      "learning_rate": 4.1e-05,
+      "loss": 0.6136,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6111054508405501,
+      "eval_loss": 0.9655206203460693,
+      "eval_runtime": 2.7965,
+      "eval_samples_per_second": 3509.691,
+      "eval_steps_per_second": 13.946,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.071944236755371,
+      "learning_rate": 4e-05,
+      "loss": 0.5745,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6065206316861946,
+      "eval_loss": 1.0082168579101562,
+      "eval_runtime": 2.8128,
+      "eval_samples_per_second": 3489.463,
+      "eval_steps_per_second": 13.865,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.7659492064952229,
+      "train_runtime": 2074.047,
+      "train_samples_per_second": 9467.047,
+      "train_steps_per_second": 36.981
+    }
+  ],
+  "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
+}