01277591c840a80e82ec8031230253b3

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

• Loss: 0.2431
• Data Size: 1.0
• Epoch Runtime: 42.0731
• Accuracy: 0.9604
• F1 Macro: 0.9584

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	1.8978	0	1.7662	0.1396	0.0827
No log	1	170	1.3900	0.0078	2.3386	0.3917	0.2318
No log	2	340	0.8690	0.0156	3.2443	0.7667	0.5991
No log	3	510	0.4160	0.0312	4.7483	0.8208	0.7037
No log	4	680	0.2093	0.0625	7.6224	0.9417	0.9224
0.0324	5	850	0.2210	0.125	10.9927	0.9333	0.9145
0.0324	6	1020	0.1457	0.25	15.8541	0.9646	0.9500
0.1835	7	1190	0.1034	0.5	26.0940	0.9708	0.9649
0.1136	8.0	1360	0.1747	1.0	44.8288	0.9604	0.9582
0.0518	9.0	1530	0.1176	1.0	44.4902	0.975	0.9786
0.0377	10.0	1700	0.2026	1.0	42.5925	0.9688	0.9651
0.0102	11.0	1870	0.2431	1.0	42.0731	0.9604	0.9584

Framework versions

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