63a7ea29ceb53332fece3a51c3afb4e1

This model is a fine-tuned version of distilbert/distilbert-base-multilingual-cased on the nyu-mll/glue [sst2] dataset. It achieves the following results on the evaluation set:

• Loss: 0.4577
• Data Size: 1.0
• Epoch Runtime: 62.9634
• Accuracy: 0.8704
• F1 Macro: 0.8698
• Rouge1: 0.8704
• Rouge2: 0.0
• Rougel: 0.8704
• Rougelsum: 0.8704

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	0.6938	0	0.8183	0.4907	0.3292	0.4907	0.0	0.4907	0.4919
No log	1	2104	0.6383	0.0078	2.0832	0.6863	0.6859	0.6875	0.0	0.6869	0.6863
No log	2	4208	0.5321	0.0156	2.1470	0.7315	0.7260	0.7326	0.0	0.7315	0.7303
0.0114	3	6312	0.4603	0.0312	3.2891	0.7766	0.7764	0.7778	0.0	0.7766	0.7755
0.4211	4	8416	0.3908	0.0625	5.3066	0.8322	0.8322	0.8322	0.0	0.8322	0.8322
0.3555	5	10520	0.4744	0.125	9.2593	0.8229	0.8212	0.8229	0.0	0.8241	0.8229
0.2608	6	12624	0.3450	0.25	16.9099	0.8542	0.8541	0.8542	0.0	0.8542	0.8542
0.2117	7	14728	0.3549	0.5	32.7095	0.8611	0.8608	0.8611	0.0	0.8611	0.8611
0.1618	8.0	16832	0.3470	1.0	63.3042	0.8657	0.8656	0.8657	0.0	0.8657	0.8657
0.1302	9.0	18936	0.4568	1.0	62.5246	0.8600	0.8599	0.8600	0.0	0.8600	0.8611
0.1067	10.0	21040	0.4577	1.0	62.9634	0.8704	0.8698	0.8704	0.0	0.8704	0.8704

Framework versions

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