3ba0289a98a5677344bb92be93321ea5

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

• Loss: 0.3801
• Data Size: 1.0
• Epoch Runtime: 55.1059
• Accuracy: 0.8935
• F1 Macro: 0.8934
• Rouge1: 0.8947
• Rouge2: 0.0
• Rougel: 0.8935
• Rougelsum: 0.8935

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	0.6993	0	0.8597	0.4919	0.3317	0.4919	0.0	0.4919	0.4931
No log	1	2104	0.5808	0.0078	1.8249	0.7963	0.7960	0.7963	0.0	0.7963	0.7963
No log	2	4208	0.4219	0.0156	1.7817	0.8056	0.8031	0.8044	0.0	0.8056	0.8056
0.0096	3	6312	0.3745	0.0312	2.7260	0.8368	0.8356	0.8368	0.0	0.8368	0.8368
0.3284	4	8416	0.2913	0.0625	4.4461	0.8692	0.8692	0.8692	0.0	0.8692	0.8692
0.2645	5	10520	0.2758	0.125	7.7715	0.8681	0.8680	0.8681	0.0	0.8681	0.8681
0.1971	6	12624	0.2813	0.25	14.3866	0.8889	0.8885	0.8889	0.0	0.8889	0.8889
0.1695	7	14728	0.3693	0.5	27.3830	0.8877	0.8871	0.8877	0.0	0.8877	0.8877
0.1588	8.0	16832	0.2904	1.0	53.1959	0.8924	0.8923	0.8924	0.0	0.8924	0.8924
0.1085	9.0	18936	0.3801	1.0	55.1059	0.8935	0.8934	0.8947	0.0	0.8935	0.8935

Framework versions

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