e2f753827efa22a90e22a049e38677c5

This model is a fine-tuned version of albert/albert-base-v1 on the nyu-mll/glue dataset. It achieves the following results on the evaluation set:

• Loss: 0.8429
• Data Size: 1.0
• Epoch Runtime: 4.8561
• Accuracy: 0.8249
• F1 Macro: 0.7995
• Rouge1: 0.8249
• Rouge2: 0.0
• Rougel: 0.8249
• Rougelsum: 0.8249

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	0.6753	0	1.1830	0.6285	0.4244	0.6291	0.0	0.6279	0.6285
No log	1	114	0.6375	0.0078	2.4839	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	2	228	0.7253	0.0156	1.3091	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	3	342	0.6300	0.0312	1.3058	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.0205	4	456	0.6197	0.0625	1.4471	0.6781	0.4543	0.6784	0.0	0.6781	0.6781
0.0205	5	570	0.6166	0.125	1.7265	0.6822	0.4647	0.6828	0.0	0.6822	0.6822
0.0205	6	684	0.4851	0.25	2.1379	0.7730	0.7284	0.7730	0.0	0.7730	0.7724
0.1315	7	798	0.4309	0.5	3.0043	0.8090	0.7734	0.8090	0.0	0.8090	0.8090
0.3474	8.0	912	0.4288	1.0	4.9612	0.8096	0.7640	0.8101	0.0	0.8090	0.8096
0.184	9.0	1026	0.4143	1.0	4.8683	0.8213	0.7978	0.8208	0.0	0.8219	0.8208
0.1427	10.0	1140	0.4761	1.0	4.8142	0.8355	0.8151	0.8355	0.0	0.8355	0.8361
0.0878	11.0	1254	0.7608	1.0	4.7622	0.8284	0.8067	0.8284	0.0	0.8290	0.8290
0.1052	12.0	1368	0.6869	1.0	4.8797	0.8296	0.7986	0.8299	0.0	0.8296	0.8296
0.0804	13.0	1482	0.8429	1.0	4.8561	0.8249	0.7995	0.8249	0.0	0.8249	0.8249

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1