d4bb5efa4446fd09c3cec4ee6b7f9f2f

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

• Loss: 0.4714
• Data Size: 1.0
• Epoch Runtime: 11.8008
• Mse: 0.4716
• Mae: 0.5129
• R2: 0.7891

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	Mse	Mae	R2
No log	0	0	5.0435	0	1.3857	5.0446	1.8525	-1.2566
No log	1	179	2.4450	0.0078	1.7986	2.4458	1.3442	-0.0941
No log	2	358	2.9948	0.0156	1.7744	2.9954	1.4071	-0.3400
No log	3	537	2.0540	0.0312	1.9364	2.0545	1.1773	0.0809
No log	4	716	1.9062	0.0625	2.3199	1.9064	1.1118	0.1472
No log	5	895	1.9882	0.125	2.8993	1.9884	1.1664	0.1105
0.0911	6	1074	1.2368	0.25	4.2179	1.2374	0.8941	0.4465
1.0993	7	1253	1.0669	0.5	6.6649	1.0674	0.8623	0.5225
0.6755	8.0	1432	0.9378	1.0	12.1193	0.9379	0.7424	0.5804
0.6614	9.0	1611	0.5567	1.0	11.7972	0.5569	0.5640	0.7509
0.4698	10.0	1790	0.4544	1.0	11.7081	0.4547	0.5174	0.7966
0.485	11.0	1969	0.5114	1.0	11.7362	0.5117	0.5589	0.7711
0.3829	12.0	2148	0.4850	1.0	11.7300	0.4853	0.5314	0.7829
0.3277	13.0	2327	0.5984	1.0	11.8775	0.5986	0.5964	0.7322
0.2764	14.0	2506	0.4714	1.0	11.8008	0.4716	0.5129	0.7891

Framework versions

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