Model save

README.md

@@ -3,8 +3,6 @@ license: mit
 base_model: indobenchmark/indogpt
 tags:
 - generated_from_trainer
-datasets:
-- id-review-gen
 model-index:
 - name: gpt2-finetune-id-review-gen
   results: []
@@ -15,9 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gpt2-finetune-id-review-gen
 
-This model is a fine-tuned version of [indobenchmark/indogpt](https://huggingface.co/indobenchmark/indogpt) on the id-review-gen dataset.
-It achieves the following results on the evaluation set:
-- Loss: 3.7620
+This model is a fine-tuned version of [indobenchmark/indogpt](https://huggingface.co/indobenchmark/indogpt) on an unknown dataset.
 
 ## Model description
 
@@ -37,23 +33,13 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 4
-- eval_batch_size: 4
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 5
-
-### Training results
-
-| Training Loss | Epoch | Step   | Validation Loss |
-|:-------------:|:-----:|:------:|:---------------:|
-| 3.9882        | 1.0   | 21065  | 3.7872          |
-| 3.6107        | 2.0   | 42130  | 3.7341          |
-| 3.3933        | 3.0   | 63195  | 3.7304          |
-| 3.2175        | 4.0   | 84260  | 3.7434          |
-| 3.0871        | 5.0   | 105325 | 3.7620          |
-
+- mixed_precision_training: Native AMP
 
 ### Framework versions