9c27487a5b00c8f5dc4e9770bafbf996

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B on the nyu-mll/glue [mrpc] dataset. It achieves the following results on the evaluation set:

• Loss: 5.3201
• Data Size: 1.0
• Epoch Runtime: 32.4833
• Accuracy: 0.7930
• F1 Macro: 0.7664
• Rouge1: 0.7930
• Rouge2: 0.0
• Rougel: 0.7936
• Rougelsum: 0.7930

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	3.8873	0	4.3244	0.5713	0.4806	0.5713	0.0	0.5713	0.5713
No log	1	114	25.9494	0.0078	5.5347	0.6645	0.4026	0.6651	0.0	0.6639	0.6645
No log	2	228	5.9864	0.0156	7.3475	0.6810	0.4797	0.6810	0.0	0.6810	0.6804
No log	3	342	8.1542	0.0312	10.3889	0.6680	0.4121	0.6686	0.0	0.6675	0.6677
0.2873	4	456	3.6128	0.0625	12.6836	0.4068	0.3817	0.4062	0.0	0.4068	0.4065
0.2873	5	570	3.3655	0.125	15.0555	0.3620	0.3050	0.3614	0.0	0.3614	0.3626
0.2873	6	684	2.5230	0.25	18.9987	0.6946	0.5443	0.6952	0.0	0.6940	0.6946
0.6552	7	798	2.2984	0.5	22.4513	0.7052	0.7030	0.7046	0.0	0.7058	0.7052
1.4743	8.0	912	2.0241	1.0	32.7387	0.7600	0.6971	0.7606	0.0	0.7600	0.7606
0.6341	9.0	1026	2.4570	1.0	33.5141	0.7683	0.7510	0.7677	0.0	0.7692	0.7689
0.5272	10.0	1140	4.2968	1.0	33.1120	0.7972	0.7572	0.7972	0.0	0.7972	0.7978
0.4274	11.0	1254	4.5681	1.0	35.1067	0.7925	0.7521	0.7925	0.0	0.7925	0.7925
0.6444	12.0	1368	5.3201	1.0	32.4833	0.7930	0.7664	0.7930	0.0	0.7936	0.7930

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.3.0
• Tokenizers 0.22.1