Huấn luyện GPT-2 hoàn tất!

README.md

@@ -0,0 +1,76 @@
+---
+library_name: transformers
+base_model: nmthien/vietnamese-gpt2
+tags:
+- generated_from_trainer
+model-index:
+- name: vietnamese-gpt2
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# vietnamese-gpt2
+
+This model is a fine-tuned version of [nmthien/vietnamese-gpt2](https://huggingface.co/nmthien/vietnamese-gpt2) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.3369
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0005
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- training_steps: 8500
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 4.5820        | 0.0588 | 500  | 4.4010          |
+| 4.2084        | 0.1176 | 1000 | 4.0623          |
+| 4.0467        | 0.1765 | 1500 | 3.8897          |
+| 3.9211        | 0.2353 | 2000 | 3.7613          |
+| 3.7690        | 0.2941 | 2500 | 3.6909          |
+| 3.7465        | 0.3529 | 3000 | 3.6145          |
+| 3.7173        | 0.4118 | 3500 | 3.5593          |
+| 3.6531        | 0.4706 | 4000 | 3.5172          |
+| 3.4956        | 0.5294 | 4500 | 3.4913          |
+| 3.4844        | 0.5882 | 5000 | 3.4561          |
+| 3.5778        | 0.6471 | 5500 | 3.4211          |
+| 3.2929        | 0.7059 | 6000 | 3.4088          |
+| 3.5503        | 0.7647 | 6500 | 3.3836          |
+| 3.5040        | 0.8235 | 7000 | 3.3714          |
+| 3.2767        | 0.8824 | 7500 | 3.3577          |
+| 3.4912        | 0.9412 | 8000 | 3.3433          |
+| 3.4930        | 1.0    | 8500 | 3.3369          |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 5.0.0
+- Tokenizers 0.22.2

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "is_local": false,
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}