DialoGPT-small-FinalFantasyDialogue

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

• Loss: 0.3930

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: 0.005
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
3.3955	1.0	141	2.8517
2.5623	1.99	282	2.1898
1.9315	3.0	424	1.7076
1.5264	4.0	565	1.3901
1.2892	4.99	706	1.1884
1.1325	6.0	848	1.0805
1.0404	7.0	989	0.9933
0.8733	8.0	1131	0.8070
0.6344	9.0	1272	0.6326
0.5047	9.99	1413	0.5504
0.413	11.0	1555	0.5021
0.3457	12.0	1696	0.4586
0.3049	12.99	1837	0.4294
0.2475	14.0	1979	0.4154
0.2081	15.0	2120	0.3943
0.1808	16.0	2262	0.3886
0.1601	17.0	2403	0.3839
0.1431	17.99	2544	0.3850
0.1323	19.0	2686	0.3843
0.1221	19.95	2820	0.3930

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3