288f4dbef4391e0d6a575b59bd32b245

This model is a fine-tuned version of Qwen/Qwen2.5-0.5B on the nyu-mll/glue [cola] dataset. It achieves the following results on the evaluation set:

• Loss: 3.8447
• Data Size: 1.0
• Epoch Runtime: 50.1274
• Accuracy: 0.6885
• F1 Macro: 0.6431
• Rouge1: 0.6875
• Rouge2: 0.0
• Rougel: 0.6895
• Rougelsum: 0.6885

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	15.8622	0	2.9769	0.3379	0.2844	0.3369	0.0	0.3379	0.3379
No log	1	267	7.2164	0.0078	4.3386	0.4678	0.4672	0.4678	0.0	0.4678	0.4678
No log	2	534	15.2909	0.0156	3.8748	0.3320	0.2747	0.3315	0.0	0.3311	0.3320
No log	3	801	2.6434	0.0312	5.0650	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
No log	4	1068	2.5827	0.0625	7.1525	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
0.2622	5	1335	2.3548	0.125	10.3028	0.6973	0.5166	0.6973	0.0	0.6973	0.6973
2.4153	6	1602	2.5367	0.25	16.4585	0.6572	0.5424	0.6562	0.0	0.6562	0.6572
2.4402	7	1869	2.9545	0.5	27.6737	0.6885	0.4078	0.6895	0.0	0.6885	0.6885
2.0841	8.0	2136	2.2678	1.0	50.5483	0.7246	0.6035	0.7246	0.0	0.7246	0.7256
1.698	9.0	2403	2.4816	1.0	49.5471	0.6855	0.5931	0.6865	0.0	0.6855	0.6851
1.4019	10.0	2670	3.0117	1.0	48.9825	0.7012	0.6256	0.7012	0.0	0.7017	0.7012
0.8822	11.0	2937	2.7553	1.0	48.3649	0.7158	0.6451	0.7158	0.0	0.7158	0.7158
0.8821	12.0	3204	3.8447	1.0	50.1274	0.6885	0.6431	0.6875	0.0	0.6895	0.6885

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.2.0
• Tokenizers 0.22.1