3934c33f25f8c3788acc6da928752df9

This model is a fine-tuned version of facebook/opt-6.7b on the nyu-mll/glue [qnli] dataset. It achieves the following results on the evaluation set:

• Loss: 0.6717
• Data Size: 1.0
• Epoch Runtime: 1416.2219
• Accuracy: 0.6062
• F1 Macro: 0.5923
• Rouge1: 0.6064
• Rouge2: 0.0
• Rougel: 0.6066
• Rougelsum: 0.6062

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	0.9384	0	15.5866	0.4950	0.4581	0.4953	0.0	0.4952	0.4952
No log	1	3273	0.7067	0.0078	26.7476	0.5053	0.3387	0.5051	0.0	0.5051	0.5053
0.0183	2	6546	0.8426	0.0156	48.7903	0.5057	0.3359	0.5053	0.0	0.5057	0.5056
0.7128	3	9819	0.6837	0.0312	75.5301	0.5461	0.5330	0.5460	0.0	0.5463	0.5460
0.6912	4	13092	0.6616	0.0625	118.7252	0.6070	0.6055	0.6067	0.0	0.6074	0.6070
0.6733	5	16365	0.6806	0.125	203.2772	0.5829	0.5383	0.5829	0.0	0.5829	0.5831
0.6802	6	19638	0.6802	0.25	372.1911	0.5778	0.5445	0.5779	0.0	0.5779	0.5779
0.647	7	22911	0.7124	0.5	718.2683	0.5640	0.5092	0.5642	0.0	0.5642	0.5639
0.6512	8.0	26184	0.6717	1.0	1416.2219	0.6062	0.5923	0.6064	0.0	0.6066	0.6062

Framework versions

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