aea982a4d96067e1bd73c9a9ea329986

This model is a fine-tuned version of Qwen/Qwen2.5-7B on the nyu-mll/glue [mrpc] dataset. It achieves the following results on the evaluation set:

• Loss: 5.6707
• Data Size: 1.0
• Epoch Runtime: 133.5309
• Accuracy: 0.6309
• F1 Macro: 0.5893
• Rouge1: 0.6309
• Rouge2: 0.0
• Rougel: 0.6315
• Rougelsum: 0.6315

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
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Data Size	Epoch Runtime	Accuracy	F1 Macro	Rouge1	Rouge2	Rougel	Rougelsum
No log	0	0	5.5142	0	5.9462	0.5926	0.5039	0.5932	0.0	0.5926	0.5932
No log	1	114	87.8027	0.0078	6.4258	0.6645	0.3992	0.6654	0.0	0.6639	0.6639
No log	2	228	26.1917	0.0156	18.3965	0.6639	0.3990	0.6645	0.0	0.6639	0.6639
No log	3	342	2.7121	0.0312	29.6723	0.6639	0.3990	0.6645	0.0	0.6639	0.6639
0.4438	4	456	22.6898	0.0625	40.0290	0.3349	0.2509	0.3343	0.0	0.3355	0.3349
0.4438	5	570	4.9299	0.125	49.3377	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.4438	6	684	3.2706	0.25	60.5316	0.3349	0.2509	0.3343	0.0	0.3355	0.3349
1.1772	7	798	2.6027	0.5	75.9534	0.6397	0.4472	0.6403	0.0	0.6397	0.6392
2.9723	8.0	912	2.6006	1.0	117.1625	0.6586	0.4727	0.6592	0.0	0.6586	0.6580
2.9076	9.0	1026	2.7372	1.0	123.9149	0.6657	0.4013	0.6663	0.0	0.6651	0.6657
2.7564	10.0	1140	2.4950	1.0	125.6964	0.6710	0.5627	0.6716	0.0	0.6704	0.6704
1.8103	11.0	1254	5.3449	1.0	122.6477	0.6132	0.5911	0.6138	0.0	0.6138	0.6138
0.9499	12.0	1368	4.8138	1.0	114.4416	0.6728	0.5441	0.6733	0.0	0.6733	0.6733
0.5053	13.0	1482	4.1730	1.0	123.2358	0.6515	0.5802	0.6512	0.0	0.6504	0.6515
0.5255	14.0	1596	5.6707	1.0	133.5309	0.6309	0.5893	0.6309	0.0	0.6315	0.6315

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.1