End of training

README.md

@@ -1,13 +1,31 @@
 ---
+language:
+- en
 base_model: Hartunka/tiny_bert_rand_50_v1
 tags:
 - generated_from_trainer
+datasets:
+- glue
 metrics:
 - accuracy
 - f1
 model-index:
 - name: tiny_bert_rand_50_v1_mrpc
-  results: []
+  results:
+  - task:
+      name: Text Classification
+      type: text-classification
+    dataset:
+      name: GLUE MRPC
+      type: glue
+      args: mrpc
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6813725490196079
+    - name: F1
+      type: f1
+      value: 0.7936507936507936
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,12 +33,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tiny_bert_rand_50_v1_mrpc
 
-This model is a fine-tuned version of [Hartunka/tiny_bert_rand_50_v1](https://huggingface.co/Hartunka/tiny_bert_rand_50_v1) on an unknown dataset.
+This model is a fine-tuned version of [Hartunka/tiny_bert_rand_50_v1](https://huggingface.co/Hartunka/tiny_bert_rand_50_v1) on the GLUE MRPC dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0427
+- Loss: 0.5949
 - Accuracy: 0.6814
-- F1: 0.7833
-- Combined Score: 0.7324
+- F1: 0.7937
+- Combined Score: 0.7375
 
 ## Model description
 

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6813725490196079,
+    "eval_combined_score": 0.7375116713352008,
+    "eval_f1": 0.7936507936507936,
+    "eval_loss": 0.5949084758758545,
+    "eval_runtime": 0.1818,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2244.026,
+    "eval_steps_per_second": 11.0,
+    "total_flos": 769504676118528.0,
+    "train_loss": 0.43935337861378987,
+    "train_runtime": 27.1726,
+    "train_samples": 3668,
+    "train_samples_per_second": 6749.449,
+    "train_steps_per_second": 27.601
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.6813725490196079,
+    "eval_combined_score": 0.7375116713352008,
+    "eval_f1": 0.7936507936507936,
+    "eval_loss": 0.5949084758758545,
+    "eval_runtime": 0.1818,
+    "eval_samples": 408,
+    "eval_samples_per_second": 2244.026,
+    "eval_steps_per_second": 11.0
+}

logs/events.out.tfevents.1744824241.s_004_m

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bde889748fc4b5d57f4de275c0ffd735fda5abfaa8ff6459ebf710400f9cb25
+size 351

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "total_flos": 769504676118528.0,
+    "train_loss": 0.43935337861378987,
+    "train_runtime": 27.1726,
+    "train_samples": 3668,
+    "train_samples_per_second": 6749.449,
+    "train_steps_per_second": 27.601
+}

trainer_state.json

@@ -0,0 +1,174 @@
+{
+  "best_metric": 0.5949084758758545,
+  "best_model_checkpoint": "tiny_bert_rand_50_v1_mrpc/checkpoint-45",
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.8691033124923706,
+      "learning_rate": 4.9e-05,
+      "loss": 0.6286,
+      "step": 15
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6862745098039216,
+      "eval_combined_score": 0.7440660474716203,
+      "eval_f1": 0.8018575851393189,
+      "eval_loss": 0.604465901851654,
+      "eval_runtime": 0.1601,
+      "eval_samples_per_second": 2547.96,
+      "eval_steps_per_second": 12.49,
+      "step": 15
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3388460874557495,
+      "learning_rate": 4.8e-05,
+      "loss": 0.5948,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6985294117647058,
+      "eval_combined_score": 0.7536192937517152,
+      "eval_f1": 0.8087091757387247,
+      "eval_loss": 0.5950304865837097,
+      "eval_runtime": 0.1669,
+      "eval_samples_per_second": 2444.764,
+      "eval_steps_per_second": 11.984,
+      "step": 30
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.4051553010940552,
+      "learning_rate": 4.7e-05,
+      "loss": 0.556,
+      "step": 45
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6813725490196079,
+      "eval_combined_score": 0.7375116713352008,
+      "eval_f1": 0.7936507936507936,
+      "eval_loss": 0.5949084758758545,
+      "eval_runtime": 0.1984,
+      "eval_samples_per_second": 2056.454,
+      "eval_steps_per_second": 10.081,
+      "step": 45
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.9397245645523071,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.5107,
+      "step": 60
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7107843137254902,
+      "eval_combined_score": 0.753316032295271,
+      "eval_f1": 0.7958477508650519,
+      "eval_loss": 0.6382691860198975,
+      "eval_runtime": 0.1676,
+      "eval_samples_per_second": 2433.718,
+      "eval_steps_per_second": 11.93,
+      "step": 60
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 3.463055372238159,
+      "learning_rate": 4.5e-05,
+      "loss": 0.4193,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6495098039215687,
+      "eval_combined_score": 0.6930790271909869,
+      "eval_f1": 0.7366482504604052,
+      "eval_loss": 0.6820002794265747,
+      "eval_runtime": 0.1579,
+      "eval_samples_per_second": 2584.681,
+      "eval_steps_per_second": 12.67,
+      "step": 75
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.737268924713135,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.3479,
+      "step": 90
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.7034313725490197,
+      "eval_combined_score": 0.7517156862745098,
+      "eval_f1": 0.8,
+      "eval_loss": 0.8077418208122253,
+      "eval_runtime": 0.1937,
+      "eval_samples_per_second": 2105.958,
+      "eval_steps_per_second": 10.323,
+      "step": 90
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.261862277984619,
+      "learning_rate": 4.3e-05,
+      "loss": 0.2647,
+      "step": 105
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.6838235294117647,
+      "eval_combined_score": 0.7316553544494722,
+      "eval_f1": 0.7794871794871795,
+      "eval_loss": 0.8842233419418335,
+      "eval_runtime": 0.1783,
+      "eval_samples_per_second": 2288.185,
+      "eval_steps_per_second": 11.217,
+      "step": 105
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 5.626375198364258,
+      "learning_rate": 4.2e-05,
+      "loss": 0.1929,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.6813725490196079,
+      "eval_combined_score": 0.7323529411764707,
+      "eval_f1": 0.7833333333333333,
+      "eval_loss": 1.042723536491394,
+      "eval_runtime": 0.1576,
+      "eval_samples_per_second": 2589.547,
+      "eval_steps_per_second": 12.694,
+      "step": 120
+    },
+    {
+      "epoch": 8.0,
+      "step": 120,
+      "total_flos": 769504676118528.0,
+      "train_loss": 0.43935337861378987,
+      "train_runtime": 27.1726,
+      "train_samples_per_second": 6749.449,
+      "train_steps_per_second": 27.601
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 750,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "total_flos": 769504676118528.0,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}