da93e95226bdf99558763b523bec045e

This model is a fine-tuned version of openai-community/gpt2-xl on the contemmcm/trec dataset. It achieves the following results on the evaluation set:

• Loss: 0.2179
• Data Size: 1.0
• Epoch Runtime: 56.9117
• Accuracy: 0.9667
• F1 Macro: 0.9687

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	1.9156	0	2.1864	0.1271	0.0808
No log	1	170	1.2179	0.0078	2.9264	0.5646	0.3946
No log	2	340	0.5068	0.0156	5.1164	0.8625	0.7181
No log	3	510	0.3445	0.0312	8.8466	0.9	0.8792
No log	4	680	0.3053	0.0625	13.3614	0.9021	0.8371
0.0329	5	850	0.1742	0.125	19.3834	0.9437	0.9416
0.0329	6	1020	0.1492	0.25	23.9653	0.9625	0.9483
0.1876	7	1190	0.1837	0.5	33.5899	0.9521	0.9585
0.1246	8.0	1360	0.1498	1.0	60.7989	0.9604	0.9346
0.0687	9.0	1530	0.1991	1.0	54.7456	0.9646	0.9672
0.0309	10.0	1700	0.2179	1.0	56.9117	0.9667	0.9687

Framework versions

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