dialochess-v3

This model is a fine-tuned version of microsoft/DialoGPT-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8843
• Accuracy: 0.0002

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3128	0.1616	200	1.1914	0.0002
1.1935	0.3231	400	1.0974	0.0002
1.1181	0.4847	600	1.0419	0.0
1.0778	0.6462	800	1.0080	0.0
1.0426	0.8078	1000	0.9828	0.0002
1.0185	0.9693	1200	0.9612	0.0002
1.0075	1.1309	1400	0.9458	0.0001
0.9765	1.2924	1600	0.9348	0.0002
0.9806	1.4540	1800	0.9248	0.0001
0.9542	1.6155	2000	0.9132	0.0002
0.9684	1.7771	2200	0.9059	0.0002
0.9525	1.9386	2400	0.9015	0.0002
0.9396	2.1002	2600	0.8960	0.0002
0.9342	2.2617	2800	0.8896	0.0002
0.9327	2.4233	3000	0.8874	0.0002
0.9344	2.5848	3200	0.8848	0.0002
0.9272	2.7464	3400	0.8848	0.0002
0.9288	2.9079	3600	0.8843	0.0002

Framework versions

• Transformers 4.57.2
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1