3d46f82bfd2431cafd42c285146e7e3c

This model is a fine-tuned version of distilbert/distilbert-base-multilingual-cased on the nyu-mll/glue [cola] dataset. It achieves the following results on the evaluation set:

• Loss: 0.9867
• Data Size: 1.0
• Epoch Runtime: 9.8389
• Accuracy: 0.7090
• F1 Macro: 0.6226
• Rouge1: 0.7090
• Rouge2: 0.0
• Rougel: 0.7090
• Rougelsum: 0.7100

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	0.7142	0	0.9053	0.3213	0.2580	0.3203	0.0	0.3213	0.3223
No log	1	267	0.6216	0.0078	1.7436	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	2	534	0.7136	0.0156	1.4116	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	3	801	0.6501	0.0312	1.7541	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	4	1068	0.6197	0.0625	1.9302	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
0.0371	5	1335	0.6244	0.125	2.3560	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
0.6008	6	1602	0.6051	0.25	3.4191	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
0.5807	7	1869	0.6334	0.5	5.4276	0.6963	0.4529	0.6973	0.0	0.6973	0.6963
0.4888	8.0	2136	0.6026	1.0	9.7041	0.7158	0.5587	0.7158	0.0	0.7158	0.7158
0.3535	9.0	2403	0.7003	1.0	9.5143	0.7168	0.6031	0.7178	0.0	0.7168	0.7168
0.2789	10.0	2670	0.8368	1.0	9.4838	0.7080	0.6271	0.7080	0.0	0.7080	0.7080
0.177	11.0	2937	1.0732	1.0	9.6678	0.7080	0.6132	0.7080	0.0	0.7080	0.7080
0.1924	12.0	3204	0.9867	1.0	9.8389	0.7090	0.6226	0.7090	0.0	0.7090	0.7100

Framework versions

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