a7bac5fd317ce8c486ea91bbc6cbce79

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct on the nyu-mll/glue [mrpc] dataset. It achieves the following results on the evaluation set:

• Loss: 8.2569
• Data Size: 1.0
• Epoch Runtime: 20.0068
• Accuracy: 0.7583
• F1 Macro: 0.6921
• Rouge1: 0.7588
• Rouge2: 0.0
• Rougel: 0.7585
• Rougelsum: 0.7583

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	6.9803	0	3.0542	0.3538	0.3020	0.3538	0.0	0.3544	0.3538
No log	1	114	47.5576	0.0078	3.5498	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
No log	2	228	19.2167	0.0156	5.2435	0.6639	0.3990	0.6645	0.0	0.6633	0.6636
No log	3	342	6.1864	0.0312	6.8450	0.6610	0.3979	0.6616	0.0	0.6604	0.6604
0.3625	4	456	2.7683	0.0625	7.9136	0.6887	0.5144	0.6887	0.0	0.6881	0.6887
0.3625	5	570	5.2220	0.125	8.8813	0.6651	0.3994	0.6657	0.0	0.6645	0.6651
0.3625	6	684	2.8104	0.25	11.1448	0.5572	0.5572	0.5572	0.0	0.5572	0.5575
0.7669	7	798	2.0181	0.5	15.0714	0.7577	0.7324	0.7571	0.0	0.7577	0.7571
1.7724	8.0	912	1.9298	1.0	23.2571	0.7695	0.7054	0.7695	0.0	0.7695	0.7695
0.6984	9.0	1026	2.5457	1.0	23.1571	0.7583	0.7051	0.7583	0.0	0.7583	0.7583
0.7234	10.0	1140	3.6471	1.0	19.5487	0.7412	0.6809	0.7417	0.0	0.7417	0.7412
0.4182	11.0	1254	3.9964	1.0	19.5287	0.7435	0.6758	0.7435	0.0	0.7435	0.7432
0.3607	12.0	1368	8.2569	1.0	20.0068	0.7583	0.6921	0.7588	0.0	0.7585	0.7583

Framework versions

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