5b71e212708b59c30c7ca90f4a5e45f9

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B on the nyu-mll/glue [sst2] dataset. It achieves the following results on the evaluation set:

• Loss: 1.5373
• Data Size: 1.0
• Epoch Runtime: 455.5306
• Accuracy: 0.8530
• F1 Macro: 0.8530
• Rouge1: 0.8530
• Rouge2: 0.0
• Rougel: 0.8530
• Rougelsum: 0.8530

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	13.7826	0	2.9984	0.5093	0.3374	0.5093	0.0	0.5093	0.5081
No log	1	2104	19.0697	0.0078	6.5043	0.5023	0.3798	0.5023	0.0	0.5023	0.5023
No log	2	4208	1.8888	0.0156	10.5801	0.7604	0.7473	0.7593	0.0	0.7604	0.7604
0.0649	3	6312	1.2229	0.0312	20.0074	0.8785	0.8782	0.8773	0.0	0.8785	0.8785
1.2665	4	8416	1.1912	0.0625	34.3975	0.8785	0.8784	0.8785	0.0	0.8796	0.8785
1.2573	5	10520	1.2551	0.125	63.8031	0.8762	0.8750	0.8762	0.0	0.8762	0.8762
0.8606	6	12624	1.1495	0.25	120.3423	0.8796	0.8795	0.8796	0.0	0.8796	0.8796
0.8488	7	14728	1.3336	0.5	231.0470	0.8773	0.8773	0.8762	0.0	0.8773	0.8773
0.6489	8.0	16832	1.3485	1.0	456.0294	0.8681	0.8681	0.8669	0.0	0.8681	0.8681
0.5127	9.0	18936	1.8534	1.0	454.2769	0.8646	0.8642	0.8646	0.0	0.8634	0.8646
0.4669	10.0	21040	1.5373	1.0	455.5306	0.8530	0.8530	0.8530	0.0	0.8530	0.8530

Framework versions

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