a358f4b173e048ae52d54a577d4b0c87

This model is a fine-tuned version of facebook/opt-350m on the nyu-mll/glue [mrpc] dataset. It achieves the following results on the evaluation set:

• Loss: 1.9134
• Data Size: 1.0
• Epoch Runtime: 20.7908
• Accuracy: 0.6728
• F1 Macro: 0.6436
• Rouge1: 0.6733
• Rouge2: 0.0
• Rougel: 0.6728
• Rougelsum: 0.6733

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	1.2504	0	3.3218	0.4298	0.4236	0.4298	0.0	0.4292	0.4298
No log	1	114	3.8409	0.0078	3.3640	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	2	228	0.8615	0.0156	3.9090	0.6887	0.5515	0.6887	0.0	0.6881	0.6881
No log	3	342	0.6389	0.0312	4.7365	0.6586	0.4004	0.6592	0.0	0.6583	0.6586
0.0414	4	456	0.6303	0.0625	5.6570	0.6769	0.5556	0.6775	0.0	0.6763	0.6769
0.0414	5	570	0.6087	0.125	7.1471	0.6857	0.4898	0.6857	0.0	0.6857	0.6857
0.0414	6	684	0.6256	0.25	9.4711	0.6828	0.4844	0.6831	0.0	0.6822	0.6828
0.1532	7	798	0.6031	0.5	13.7562	0.6899	0.6093	0.6899	0.0	0.6893	0.6893
0.5617	8.0	912	0.5837	1.0	21.3417	0.6851	0.4907	0.6857	0.0	0.6846	0.6851
0.3501	9.0	1026	0.5776	1.0	20.5980	0.7488	0.7065	0.7488	0.0	0.7494	0.7494
0.1945	10.0	1140	0.8856	1.0	20.4729	0.6893	0.6768	0.6893	0.0	0.6904	0.6899
0.1308	11.0	1254	1.0776	1.0	21.9111	0.6733	0.6524	0.6733	0.0	0.6739	0.6739
0.1421	12.0	1368	1.4173	1.0	21.0113	0.7258	0.6760	0.7252	0.0	0.7258	0.7264
0.0937	13.0	1482	1.9134	1.0	20.7908	0.6728	0.6436	0.6733	0.0	0.6728	0.6733

Framework versions

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