94d27c830f21e0d3b965e5f79dfb3a4b

This model is a fine-tuned version of google-bert/bert-large-cased-whole-word-masking-finetuned-squad on the nyu-mll/glue [mnli] dataset. It achieves the following results on the evaluation set:

• Loss: 1.0991
• Data Size: 1.0
• Epoch Runtime: 1123.8548
• Accuracy: 0.3182
• F1 Macro: 0.1609
• Rouge1: 0.3184
• Rouge2: 0.0
• Rougel: 0.3182
• Rougelsum: 0.3183

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	1.2096	0	7.9319	0.3545	0.1745	0.3544	0.0	0.3545	0.3543
1.0646	1	12271	0.8609	0.0078	17.8166	0.6177	0.6030	0.6175	0.0	0.6180	0.6179
0.7579	2	24542	0.6848	0.0156	27.5338	0.7223	0.7156	0.7223	0.0	0.7226	0.7226
0.7048	3	36813	0.6396	0.0312	44.3246	0.7515	0.7499	0.7511	0.0	0.7519	0.7517
0.6735	4	49084	0.6076	0.0625	79.3339	0.7682	0.7669	0.7680	0.0	0.7684	0.7685
0.6336	5	61355	0.6350	0.125	148.9773	0.7508	0.7401	0.7506	0.0	0.7509	0.7509
0.6624	6	73626	0.6155	0.25	290.3250	0.7625	0.7623	0.7624	0.0	0.7628	0.7626
0.6988	7	85897	0.6640	0.5	568.7869	0.7384	0.7395	0.7383	0.0	0.7385	0.7385
1.1057	8.0	98168	1.0991	1.0	1123.8548	0.3182	0.1609	0.3184	0.0	0.3182	0.3183

Framework versions

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