9b0c92f90d7aa56be8b0bb3e1acda1dc

This model is a fine-tuned version of albert/albert-xxlarge-v1 on the nyu-mll/glue dataset. It achieves the following results on the evaluation set:

• Loss: 0.6089
• Data Size: 1.0
• Epoch Runtime: 11.4404
• Accuracy: 0.8874
• F1 Macro: 0.8746
• Rouge1: 0.8880
• Rouge2: 0.0
• Rougel: 0.8874
• Rougelsum: 0.8874

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	0.6472	0	1.6991	0.6675	0.4103	0.6680	0.0	0.6669	0.6675
No log	1	114	0.6599	0.0078	1.9507	0.6633	0.4196	0.6639	0.0	0.6633	0.6633
No log	2	228	0.6848	0.0156	2.0740	0.6639	0.4105	0.6645	0.0	0.6633	0.6645
No log	3	342	0.6442	0.0312	2.3715	0.6645	0.4958	0.6645	0.0	0.6645	0.6645
0.0213	4	456	0.6389	0.0625	2.7245	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.0213	5	570	0.6604	0.125	3.2922	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.0213	6	684	0.5255	0.25	4.4586	0.7647	0.6954	0.7642	0.0	0.7653	0.7653
0.1444	7	798	0.3578	0.5	6.7720	0.8785	0.8591	0.8785	0.0	0.8785	0.8791
0.3439	8.0	912	0.2912	1.0	11.6550	0.8868	0.8735	0.8868	0.0	0.8874	0.8868
0.1655	9.0	1026	0.3735	1.0	11.4332	0.8697	0.8554	0.8703	0.0	0.8697	0.8697
0.0987	10.0	1140	0.5588	1.0	11.4799	0.8809	0.8604	0.8809	0.0	0.8809	0.8809
0.0641	11.0	1254	0.6764	1.0	11.4494	0.8809	0.8661	0.8812	0.0	0.8809	0.8815
0.0563	12.0	1368	0.6089	1.0	11.4404	0.8874	0.8746	0.8880	0.0	0.8874	0.8874

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1