End of training

README.md

@@ -1,13 +1,28 @@
 ---
 library_name: transformers
+language:
+- en
 base_model: Hartunka/tiny_bert_rand_5_v2
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 model-index:
 - name: tiny_bert_rand_5_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: 0.6216436126932465
 ---
 
 <!-- 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_5_v2_mnli
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_rand_5_v2](https://huggingface.co/Hartunka/tiny_bert_rand_5_v2) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_rand_5_v2](https://huggingface.co/Hartunka/tiny_bert_rand_5_v2) on the GLUE MNLI dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9851
-- Accuracy: 0.6293
+- Loss: 0.8521
+- Accuracy: 0.6216
 
 ## Model description
 

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6164034640855833,
+    "eval_accuracy_mm": 0.6216436126932465,
+    "eval_loss": 0.8499380350112915,
+    "eval_loss_mm": 0.8520750999450684,
+    "eval_runtime": 2.8358,
+    "eval_runtime_mm": 2.8349,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3461.094,
+    "eval_samples_per_second_mm": 3468.252,
+    "eval_steps_per_second": 13.753,
+    "eval_steps_per_second_mm": 13.757,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7496284455022103,
+    "train_runtime": 2069.0979,
+    "train_samples": 392702,
+    "train_samples_per_second": 9489.691,
+    "train_steps_per_second": 37.069
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "epoch_mm": 10.0,
+    "eval_accuracy": 0.6164034640855833,
+    "eval_accuracy_mm": 0.6216436126932465,
+    "eval_loss": 0.8499380350112915,
+    "eval_loss_mm": 0.8520750999450684,
+    "eval_runtime": 2.8358,
+    "eval_runtime_mm": 2.8349,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 3461.094,
+    "eval_samples_per_second_mm": 3468.252,
+    "eval_steps_per_second": 13.753,
+    "eval_steps_per_second_mm": 13.757
+}

logs/events.out.tfevents.1745297654.s_005_m.2836625.107

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2bcbe6093f2b2c629164d964c791d65b3b3206312357b17fd9b0d3e9f59a247
+size 686

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1.029836100834816e+17,
+    "train_loss": 0.7496284455022103,
+    "train_runtime": 2069.0979,
+    "train_samples": 392702,
+    "train_samples_per_second": 9489.691,
+    "train_steps_per_second": 37.069
+}

trainer_state.json

@@ -0,0 +1,212 @@
+{
+  "best_global_step": 7670,
+  "best_metric": 0.8499380350112915,
+  "best_model_checkpoint": "tiny_bert_rand_5_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": 1.1199190616607666,
+      "learning_rate": 4.9e-05,
+      "loss": 0.9894,
+      "step": 1534
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5552725420275089,
+      "eval_loss": 0.9260166883468628,
+      "eval_runtime": 2.8587,
+      "eval_samples_per_second": 3433.368,
+      "eval_steps_per_second": 13.643,
+      "step": 1534
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.4727002382278442,
+      "learning_rate": 4.8e-05,
+      "loss": 0.9044,
+      "step": 3068
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5842078451349975,
+      "eval_loss": 0.8912194967269897,
+      "eval_runtime": 2.8033,
+      "eval_samples_per_second": 3501.241,
+      "eval_steps_per_second": 13.912,
+      "step": 3068
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.6097134351730347,
+      "learning_rate": 4.7e-05,
+      "loss": 0.8553,
+      "step": 4602
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5969434538970962,
+      "eval_loss": 0.8659091591835022,
+      "eval_runtime": 2.8097,
+      "eval_samples_per_second": 3493.315,
+      "eval_steps_per_second": 13.881,
+      "step": 4602
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.6552950143814087,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.8096,
+      "step": 6136
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.60825267447784,
+      "eval_loss": 0.8656396269798279,
+      "eval_runtime": 2.7768,
+      "eval_samples_per_second": 3534.651,
+      "eval_steps_per_second": 14.045,
+      "step": 6136
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.7045304775238037,
+      "learning_rate": 4.5e-05,
+      "loss": 0.765,
+      "step": 7670
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6164034640855833,
+      "eval_loss": 0.8499380350112915,
+      "eval_runtime": 2.8047,
+      "eval_samples_per_second": 3499.494,
+      "eval_steps_per_second": 13.905,
+      "step": 7670
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.182920217514038,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7195,
+      "step": 9204
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6299541518084565,
+      "eval_loss": 0.8635302186012268,
+      "eval_runtime": 2.7851,
+      "eval_samples_per_second": 3524.115,
+      "eval_steps_per_second": 14.003,
+      "step": 9204
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 2.068575859069824,
+      "learning_rate": 4.3e-05,
+      "loss": 0.6754,
+      "step": 10738
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6362710137544575,
+      "eval_loss": 0.8769534230232239,
+      "eval_runtime": 2.8206,
+      "eval_samples_per_second": 3479.748,
+      "eval_steps_per_second": 13.827,
+      "step": 10738
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.3491876125335693,
+      "learning_rate": 4.2e-05,
+      "loss": 0.6336,
+      "step": 12272
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6287315333672949,
+      "eval_loss": 0.914506733417511,
+      "eval_runtime": 2.8023,
+      "eval_samples_per_second": 3502.433,
+      "eval_steps_per_second": 13.917,
+      "step": 12272
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 3.0106260776519775,
+      "learning_rate": 4.1e-05,
+      "loss": 0.5919,
+      "step": 13806
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.6296484971981661,
+      "eval_loss": 0.9488242268562317,
+      "eval_runtime": 2.8102,
+      "eval_samples_per_second": 3492.655,
+      "eval_steps_per_second": 13.878,
+      "step": 13806
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.469269275665283,
+      "learning_rate": 4e-05,
+      "loss": 0.5523,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.6293428425878757,
+      "eval_loss": 0.9850627183914185,
+      "eval_runtime": 2.7995,
+      "eval_samples_per_second": 3506.037,
+      "eval_steps_per_second": 13.931,
+      "step": 15340
+    },
+    {
+      "epoch": 10.0,
+      "step": 15340,
+      "total_flos": 1.029836100834816e+17,
+      "train_loss": 0.7496284455022103,
+      "train_runtime": 2069.0979,
+      "train_samples_per_second": 9489.691,
+      "train_steps_per_second": 37.069
+    }
+  ],
+  "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
+}