1cc28775a62dd91a8453ed5fa0b05cd2

This model is a fine-tuned version of openai-community/gpt2 on the nyu-mll/glue [wnli] dataset. It achieves the following results on the evaluation set:

• Loss: 0.7420
• Data Size: 1.0
• Epoch Runtime: 2.9515
• Accuracy: 0.4844
• F1 Macro: 0.3263
• Rouge1: 0.4844
• Rouge2: 0.0
• Rougel: 0.4844
• Rougelsum: 0.4844

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	6.1621	0	0.6725	0.5625	0.36	0.5625	0.0	0.5625	0.5625
No log	1	19	2.1010	0.0078	1.2433	0.5625	0.4589	0.5625	0.0	0.5625	0.5625
No log	2	38	3.0031	0.0156	1.0712	0.4531	0.4003	0.4531	0.0	0.4531	0.4531
No log	3	57	1.9434	0.0312	1.1427	0.4688	0.4333	0.4688	0.0	0.4688	0.4688
No log	4	76	1.4283	0.0625	1.3860	0.5938	0.4340	0.5938	0.0	0.5938	0.5938
No log	5	95	0.9721	0.125	1.5890	0.5156	0.4688	0.5156	0.0	0.5156	0.5156
0.2654	6	114	0.8514	0.25	1.6045	0.5312	0.4203	0.5312	0.0	0.5312	0.5312
0.2654	7	133	0.7629	0.5	2.0730	0.375	0.3695	0.375	0.0	0.375	0.375
0.607	8.0	152	0.7792	1.0	2.9196	0.4844	0.3915	0.4844	0.0	0.4844	0.4844
0.607	9.0	171	0.7194	1.0	2.5582	0.5312	0.375	0.5312	0.0	0.5312	0.5312
0.607	10.0	190	0.7205	1.0	2.6787	0.3281	0.3200	0.3281	0.0	0.3281	0.3281
0.7217	11.0	209	0.7661	1.0	2.8739	0.4531	0.3347	0.4531	0.0	0.4531	0.4531
0.7217	12.0	228	0.7355	1.0	3.1090	0.5312	0.3469	0.5312	0.0	0.5312	0.5312
0.7217	13.0	247	0.7420	1.0	2.9515	0.4844	0.3263	0.4844	0.0	0.4844	0.4844

Framework versions

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