efb6876f57e9f860ae8e741bc97bfe9e

This model is a fine-tuned version of google-bert/bert-base-german-dbmdz-uncased on the nyu-mll/glue [qnli] dataset. It achieves the following results on the evaluation set:

• Loss: 0.7024
• Data Size: 0.5
• Epoch Runtime: 77.6977
• Accuracy: 0.4943
• F1 Macro: 0.3308
• Rouge1: 0.4947
• Rouge2: 0.0
• Rougel: 0.4943
• Rougelsum: 0.4944

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	0.8479	0	2.8400	0.4932	0.3341	0.4934	0.0	0.4932	0.4934
No log	1	3273	0.6964	0.0078	4.2657	0.5057	0.3359	0.5053	0.0	0.5057	0.5056
0.0114	2	6546	0.6932	0.0156	5.4455	0.5057	0.3359	0.5053	0.0	0.5057	0.5056
0.7138	3	9819	0.6931	0.0312	7.7829	0.5057	0.3359	0.5053	0.0	0.5057	0.5056
0.7044	4	13092	0.6939	0.0625	12.5824	0.4943	0.3308	0.4947	0.0	0.4943	0.4944
0.7084	5	16365	0.7004	0.125	21.8616	0.4943	0.3308	0.4947	0.0	0.4943	0.4944
0.699	6	19638	0.6950	0.25	40.8653	0.4943	0.3308	0.4947	0.0	0.4943	0.4944
0.7087	7	22911	0.7024	0.5	77.6977	0.4943	0.3308	0.4947	0.0	0.4943	0.4944

Framework versions

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