a4ffa8d70a73b18fee2fd0f21c309a70

This model is a fine-tuned version of Qwen/Qwen2.5-3B on the nyu-mll/glue [mrpc] dataset. It achieves the following results on the evaluation set:

• Loss: 9.9590
• Data Size: 1.0
• Epoch Runtime: 48.0043
• Accuracy: 0.6474
• F1 Macro: 0.5715
• Rouge1: 0.6474
• Rouge2: 0.0
• Rougel: 0.6468
• Rougelsum: 0.6474

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	4.3212	0	5.7017	0.6604	0.4480	0.6610	0.0	0.6598	0.6598
No log	1	114	74.6144	0.0078	5.7210	0.3349	0.2529	0.3343	0.0	0.3355	0.3349
No log	2	228	14.3921	0.0156	9.3497	0.6680	0.4262	0.6686	0.0	0.6680	0.6680
No log	3	342	12.7665	0.0312	14.4486	0.3414	0.2704	0.3408	0.0	0.3417	0.3414
0.7218	4	456	8.6652	0.0625	18.3708	0.4198	0.4118	0.4198	0.0	0.4192	0.4198
0.7218	5	570	2.5875	0.125	22.9430	0.6621	0.4000	0.6627	0.0	0.6619	0.6621
0.7218	6	684	3.0767	0.25	25.4013	0.6657	0.4013	0.6663	0.0	0.6654	0.6654
0.7678	7	798	3.6529	0.5	34.2292	0.3343	0.2519	0.3337	0.0	0.3349	0.3346
2.7535	8.0	912	3.5080	1.0	47.6503	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
2.8448	9.0	1026	2.5842	1.0	49.2224	0.6645	0.3992	0.6651	0.0	0.6645	0.6645
2.741	10.0	1140	2.6351	1.0	43.8692	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
2.6939	11.0	1254	2.5636	1.0	48.5788	0.6816	0.5249	0.6825	0.0	0.6810	0.6816
2.5364	12.0	1368	2.9769	1.0	44.7631	0.6492	0.5980	0.6489	0.0	0.6498	0.6492
1.2337	13.0	1482	3.1982	1.0	50.8832	0.6533	0.5766	0.6533	0.0	0.6521	0.6533
0.7026	14.0	1596	4.9728	1.0	44.5094	0.6114	0.5733	0.6108	0.0	0.6108	0.6114
1.5571	15.0	1710	9.9590	1.0	48.0043	0.6474	0.5715	0.6474	0.0	0.6468	0.6474

Framework versions

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