ea4f39a3aff252fdc48673c9fc969e6e

This model is a fine-tuned version of albert/albert-xlarge-v2 on the contemmcm/trec dataset. It achieves the following results on the evaluation set:

• Loss: 1.6620
• Data Size: 1.0
• Epoch Runtime: 12.0073
• Accuracy: 0.2771
• F1 Macro: 0.0723

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
No log	0	0	1.8786	0	0.8803	0.1646	0.0471
No log	1	170	1.8196	0.0078	1.1461	0.1792	0.0506
No log	2	340	1.7277	0.0156	1.2108	0.2771	0.0723
No log	3	510	1.7766	0.0312	1.4517	0.1792	0.0506
No log	4	680	1.6518	0.0625	1.7072	0.2771	0.0723
0.1031	5	850	1.7210	0.125	2.4162	0.1792	0.0506
0.1031	6	1020	1.7100	0.25	3.7895	0.1792	0.0506
1.6747	7	1190	1.7451	0.5	6.5929	0.1333	0.0392
1.6733	8.0	1360	1.6620	1.0	12.0073	0.2771	0.0723

Framework versions

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