1a7d904346e7c4adfb1de13434e74811

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

• Loss: 0.8393
• Data Size: 1.0
• Epoch Runtime: 11.5957
• Accuracy: 0.7365
• F1 Macro: 0.7092
• Rouge1: 0.7365
• Rouge2: 0.0
• Rougel: 0.7359
• Rougelsum: 0.7362

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	0.7204	0	1.6779	0.4779	0.4482	0.4779	0.0	0.4782	0.4773
No log	1	294	0.7127	0.0078	3.0497	0.4605	0.4594	0.4602	0.0	0.4608	0.4602
No log	2	588	0.6631	0.0156	1.8627	0.6213	0.3832	0.6213	0.0	0.6207	0.6210
No log	3	882	0.6623	0.0312	2.0432	0.6222	0.4283	0.6222	0.0	0.6213	0.6225
0.0272	4	1176	0.6547	0.0625	2.3096	0.6219	0.3850	0.6219	0.0	0.6215	0.6216
0.0546	5	1470	0.6508	0.125	2.9116	0.6385	0.4884	0.6382	0.0	0.6382	0.6385
0.0898	6	1764	0.6159	0.25	4.1202	0.6615	0.5994	0.6615	0.0	0.6612	0.6615
0.5742	7	2058	0.5887	0.5	6.5760	0.6893	0.6685	0.6890	0.0	0.6893	0.6893
0.5151	8.0	2352	0.5455	1.0	11.4274	0.7230	0.6932	0.7237	0.0	0.7233	0.7233
0.4331	9.0	2646	0.5818	1.0	11.3375	0.7307	0.7003	0.7310	0.0	0.7304	0.7307
0.2983	10.0	2940	0.6985	1.0	11.3204	0.7374	0.7160	0.7381	0.0	0.7374	0.7374
0.2542	11.0	3234	0.7486	1.0	11.4927	0.7405	0.7218	0.7405	0.0	0.7405	0.7405
0.1905	12.0	3528	0.8393	1.0	11.5957	0.7365	0.7092	0.7365	0.0	0.7359	0.7362

Framework versions

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