17c31f5eac76c65ecd9e437c56877227

This model is a fine-tuned version of Qwen/Qwen2.5-7B on the nyu-mll/glue [mnli] dataset. It achieves the following results on the evaluation set:

• Loss: 5.0623
• Data Size: 1.0
• Epoch Runtime: 9754.0753
• Accuracy: 0.5706
• F1 Macro: 0.5657
• Rouge1: 0.5710
• Rouge2: 0.0
• Rougel: 0.5709
• Rougelsum: 0.5706

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	27.0052	0	28.9664	0.3176	0.2542	0.3177	0.0	0.3175	0.3174
11.5036	1	12271	4.8379	0.0078	103.5636	0.3186	0.1653	0.3189	0.0	0.3185	0.3187
7.3197	2	24542	4.7337	0.0156	197.2104	0.3500	0.2225	0.3499	0.0	0.3500	0.3499
4.8706	3	36813	4.4903	0.0312	348.8228	0.3216	0.2579	0.3217	0.0	0.3217	0.3217
4.5381	4	49084	4.6209	0.0625	661.1492	0.3400	0.2550	0.3398	0.0	0.3398	0.3399
4.4433	5	61355	4.4599	0.125	1299.6591	0.3354	0.2520	0.3355	0.0	0.3351	0.3354
4.4624	6	73626	4.5043	0.25	2539.0743	0.3309	0.2632	0.3311	0.0	0.3309	0.3309
4.4128	7	85897	4.4061	0.5	5071.7366	0.3281	0.1708	0.3281	0.0	0.3282	0.3283
3.7251	8.0	98168	3.6796	1.0	10036.8718	0.5636	0.5641	0.5635	0.0	0.5635	0.5638
3.5089	9.0	110439	3.5933	1.0	10113.0923	0.5770	0.5776	0.5770	0.0	0.5769	0.5770
3.1793	10.0	122710	3.5824	1.0	10057.9050	0.5976	0.5941	0.5977	0.0	0.5977	0.5976
2.995	11.0	134981	3.6873	1.0	9700.8608	0.5871	0.5883	0.5872	0.0	0.5870	0.5878
2.6179	12.0	147252	3.8272	1.0	9767.2053	0.5911	0.5911	0.5913	0.0	0.5912	0.5911
2.1335	13.0	159523	4.5428	1.0	9732.3295	0.5827	0.5820	0.5828	0.0	0.5827	0.5825
1.7439	14.0	171794	5.0623	1.0	9754.0753	0.5706	0.5657	0.5710	0.0	0.5709	0.5706

Framework versions

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