05fe6dd9fb736f262d9c7b87e3d62b20

This model is a fine-tuned version of google-bert/bert-large-cased-whole-word-masking on the dim/tldr_news dataset. It achieves the following results on the evaluation set:

• Loss: 1.2059
• Data Size: 1.0
• Epoch Runtime: 19.6449
• Accuracy: 0.7812
• F1 Macro: 0.8166
• Rouge1: 0.7820
• Rouge2: 0.0
• Rougel: 0.7820
• Rougelsum: 0.7812

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	1.6881	0	1.6333	0.2038	0.1709	0.2031	0.0	0.2038	0.2031
No log	1	178	1.6149	0.0078	2.6729	0.2152	0.0743	0.2145	0.0	0.2152	0.2152
No log	2	356	1.4364	0.0156	2.2857	0.2401	0.0774	0.2401	0.0	0.2408	0.2393
No log	3	534	0.9107	0.0312	3.1075	0.6349	0.4959	0.6342	0.0	0.6349	0.6349
No log	4	712	0.9184	0.0625	4.2252	0.6832	0.5360	0.6832	0.0	0.6839	0.6832
No log	5	890	0.7633	0.125	5.6194	0.7088	0.5466	0.7095	0.0	0.7088	0.7095
0.0548	6	1068	0.7143	0.25	8.3142	0.7131	0.5727	0.7138	0.0	0.7138	0.7138
0.5851	7	1246	0.6713	0.5	11.8549	0.7514	0.7612	0.7521	0.0	0.7514	0.7514
0.4653	8.0	1424	0.6280	1.0	20.6589	0.7550	0.7857	0.7557	0.0	0.7550	0.7550
0.2572	9.0	1602	0.7753	1.0	20.7344	0.7678	0.7883	0.7685	0.0	0.7678	0.7678
0.1493	10.0	1780	0.9711	1.0	20.1288	0.7486	0.7886	0.7493	0.0	0.7493	0.7493
0.097	11.0	1958	1.1534	1.0	20.0724	0.7578	0.7986	0.7585	0.0	0.7585	0.7578
0.103	12.0	2136	1.2059	1.0	19.6449	0.7812	0.8166	0.7820	0.0	0.7820	0.7812

Framework versions

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