game_music_gen_V06

This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9341

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: 8
• eval_batch_size: 4
• seed: 1
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
2.3569	2.7027	500	1.8166
1.623	5.4054	1000	1.5719
1.3792	8.1081	1500	1.4346
1.2156	10.8108	2000	1.3726
1.0928	13.5135	2500	1.2929
0.9815	16.2162	3000	1.2165
0.8874	18.9189	3500	1.1643
0.7956	21.6216	4000	1.1065
0.7028	24.3243	4500	1.0516
0.6261	27.0270	5000	1.0120
0.5457	29.7297	5500	0.9647
0.4771	32.4324	6000	0.9561
0.4184	35.1351	6500	0.9345
0.3755	37.8378	7000	0.9279
0.3372	40.5405	7500	0.9354
0.3182	43.2432	8000	0.9319
0.3011	45.9459	8500	0.9352
0.2972	48.6486	9000	0.9341

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0