| | --- |
| | library_name: transformers |
| | base_model: /fs-computility/plm/linzhouhan/daibeiya/models/gpt2-large |
| | tags: |
| | - generated_from_trainer |
| | datasets: |
| | - openwebtext |
| | model-index: |
| | - name: gpt2_large_contextlm_l0236_add_lnnorm_lr_bf16_lr6e-4 |
| | 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. --> |
| |
|
| | # gpt2_large_contextlm_l0236_add_lnnorm_lr_bf16_lr6e-4 |
| |
|
| | This model is a fine-tuned version of [/fs-computility/plm/linzhouhan/daibeiya/models/gpt2-large](https://huggingface.co//fs-computility/plm/linzhouhan/daibeiya/models/gpt2-large) on the openwebtext dataset. |
| | It achieves the following results on the evaluation set: |
| | - Loss: 2.7357 |
| |
|
| | ## 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: 0.0006 |
| | - train_batch_size: 4 |
| | - eval_batch_size: 8 |
| | - seed: 42 |
| | - distributed_type: multi-GPU |
| | - num_devices: 64 |
| | - gradient_accumulation_steps: 2 |
| | - total_train_batch_size: 512 |
| | - total_eval_batch_size: 512 |
| | - optimizer: Use adamw_torch with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
| | - lr_scheduler_type: cosine |
| | - lr_scheduler_warmup_ratio: 0.05 |
| | - num_epochs: 1.0 |
| | |
| | ### Training results |
| | |
| | | Training Loss | Epoch | Step | Validation Loss | |
| | |:-------------:|:------:|:-----:|:---------------:| |
| | | 3.8376 | 0.0580 | 1000 | 3.7978 | |
| | | 3.3595 | 0.1160 | 2000 | 3.3179 | |
| | | 3.207 | 0.1741 | 3000 | 3.1580 | |
| | | 3.0999 | 0.2321 | 4000 | 3.0631 | |
| | | 3.0333 | 0.2901 | 5000 | 2.9981 | |
| | | 2.9958 | 0.3481 | 6000 | 2.9496 | |
| | | 2.9443 | 0.4062 | 7000 | 2.9102 | |
| | | 2.9097 | 0.4642 | 8000 | 2.8760 | |
| | | 2.879 | 0.5222 | 9000 | 2.8451 | |
| | | 2.8506 | 0.5802 | 10000 | 2.8198 | |
| | | 2.831 | 0.6383 | 11000 | 2.7969 | |
| | | 2.8156 | 0.6963 | 12000 | 2.7781 | |
| | | 2.799 | 0.7543 | 13000 | 2.7616 | |
| | | 2.7802 | 0.8123 | 14000 | 2.7494 | |
| | | 2.7785 | 0.8703 | 15000 | 2.7414 | |
| | | 2.7706 | 0.9284 | 16000 | 2.7370 | |
| | | 2.7665 | 0.9864 | 17000 | 2.7357 | |
| | |
| | |
| | ### Framework versions |
| | |
| | - Transformers 4.51.3 |
| | - Pytorch 2.3.0+cu121 |
| | - Datasets 4.0.0 |
| | - Tokenizers 0.21.4 |
| | |
