End of training

README.md

@@ -14,8 +14,6 @@ should probably proofread and complete it, then remove this comment. -->
 # bert-base-banking77-pt2
 
 This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 1.7084
 
 ## Model description
 
@@ -35,24 +33,17 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 1
-- eval_batch_size: 8
+- train_batch_size: 32
+- eval_batch_size: 32
 - seed: 42
+- gradient_accumulation_steps: 20
+- total_train_batch_size: 640
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 7
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 16.994        | 1.0   | 10   | 4.2000          |
-| 3.8266        | 2.0   | 20   | 3.1920          |
-| 3.0646        | 3.0   | 30   | 2.2112          |
-| 2.3667        | 4.0   | 40   | 1.9768          |
-| 2.0815        | 5.0   | 50   | 1.8096          |
-| 1.9727        | 6.0   | 60   | 1.7462          |
-| 1.8305        | 7.0   | 70   | 1.7084          |
 
 
 ### Framework versions

logs/events.out.tfevents.1702224308.a2c88cfea1e4.588.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d951358e05c804da4cbbca25400d25cc6154bed5bb666c8cc60609ca655774b
-size 6134
+oid sha256:1dbe5dcff01b4fc9abfd8ce67e62c005d3562f64b797cda1e32abca33554c47b
+size 6288

logs/events.out.tfevents.1702224578.a2c88cfea1e4.588.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:597d92f5bc0ad1cc956dd20900424d45d5c6b6870be86287e923f511d85e1567
+size 4803

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95570e5a301362c930fde7b10ee66a21c8f42e3c240dc4cf312d40a02bdcfb5d
+oid sha256:fbdf21c65e42ecf425af13ee6cb1bc584fb68ab844200b542e0174b7ee5e4cb4
 size 497814144

trainer_state.json

@@ -1,126 +1,28 @@
 {
-  "best_metric": 1.7083709239959717,
-  "best_model_checkpoint": "bert-base-banking77-pt2/checkpoint-70",
-  "epoch": 7.0,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.375,
   "eval_steps": 10,
-  "global_step": 70,
+  "global_step": 7,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "learning_rate": 4.2857142857142856e-05,
-      "loss": 16.994,
-      "step": 10
-    },
-    {
-      "epoch": 1.0,
-      "eval_loss": 4.199960231781006,
-      "eval_runtime": 0.1251,
-      "eval_samples_per_second": 79.92,
-      "eval_steps_per_second": 15.984,
-      "step": 10
-    },
-    {
-      "epoch": 2.0,
-      "learning_rate": 3.571428571428572e-05,
-      "loss": 3.8266,
-      "step": 20
-    },
-    {
-      "epoch": 2.0,
-      "eval_loss": 3.1919631958007812,
-      "eval_runtime": 0.1628,
-      "eval_samples_per_second": 61.412,
-      "eval_steps_per_second": 12.282,
-      "step": 20
-    },
-    {
-      "epoch": 3.0,
-      "learning_rate": 2.857142857142857e-05,
-      "loss": 3.0646,
-      "step": 30
-    },
-    {
-      "epoch": 3.0,
-      "eval_loss": 2.2111892700195312,
-      "eval_runtime": 0.1271,
-      "eval_samples_per_second": 78.703,
-      "eval_steps_per_second": 15.741,
-      "step": 30
-    },
-    {
-      "epoch": 4.0,
-      "learning_rate": 2.1428571428571428e-05,
-      "loss": 2.3667,
-      "step": 40
-    },
-    {
-      "epoch": 4.0,
-      "eval_loss": 1.9768412113189697,
-      "eval_runtime": 0.1291,
-      "eval_samples_per_second": 77.443,
-      "eval_steps_per_second": 15.489,
-      "step": 40
-    },
-    {
-      "epoch": 5.0,
-      "learning_rate": 1.4285714285714285e-05,
-      "loss": 2.0815,
-      "step": 50
-    },
-    {
-      "epoch": 5.0,
-      "eval_loss": 1.8095529079437256,
-      "eval_runtime": 0.1244,
-      "eval_samples_per_second": 80.397,
-      "eval_steps_per_second": 16.079,
-      "step": 50
-    },
-    {
-      "epoch": 6.0,
-      "learning_rate": 7.142857142857143e-06,
-      "loss": 1.9727,
-      "step": 60
-    },
-    {
-      "epoch": 6.0,
-      "eval_loss": 1.7462323904037476,
-      "eval_runtime": 0.1467,
-      "eval_samples_per_second": 68.173,
-      "eval_steps_per_second": 13.635,
-      "step": 60
-    },
-    {
-      "epoch": 7.0,
-      "learning_rate": 0.0,
-      "loss": 1.8305,
-      "step": 70
-    },
-    {
-      "epoch": 7.0,
-      "eval_loss": 1.7083709239959717,
-      "eval_runtime": 0.1559,
-      "eval_samples_per_second": 64.127,
-      "eval_steps_per_second": 12.825,
-      "step": 70
-    },
-    {
-      "epoch": 7.0,
-      "step": 70,
-      "total_flos": 4572610560000.0,
-      "train_loss": 4.5909277507237025,
-      "train_runtime": 92.0015,
-      "train_samples_per_second": 0.761,
-      "train_steps_per_second": 0.761
+      "epoch": 4.38,
+      "step": 7,
+      "total_flos": 286376067072000.0,
+      "train_loss": 23.388562883649552,
+      "train_runtime": 165.8541,
+      "train_samples_per_second": 42.206,
+      "train_steps_per_second": 0.042
     }
   ],
   "logging_steps": 10,
-  "max_steps": 70,
+  "max_steps": 7,
   "num_train_epochs": 7,
   "save_steps": 10,
-  "total_flos": 4572610560000.0,
+  "total_flos": 286376067072000.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:927eb5b40dd0e5373dd84ce242021c3db8b3eefba4ad8144befc58a13e51b4a7
+oid sha256:3120a79b950bf9a79bef5fab066d940b2c8afce2622e34f82e0c5edb68126598
 size 4600