95f5641a17f08d66add210ca06acebd7

This model is a fine-tuned version of Qwen/Qwen2.5-0.5B on the nyu-mll/glue [sst2] dataset. It achieves the following results on the evaluation set:

• Loss: 3.1289
• Data Size: 1.0
• Epoch Runtime: 353.5838
• Accuracy: 0.8194
• F1 Macro: 0.8193
• Rouge1: 0.8194
• Rouge2: 0.0
• Rougel: 0.8183
• Rougelsum: 0.8194

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	7.6262	0	2.7700	0.5046	0.3963	0.5058	0.0	0.5046	0.5035
No log	1	2104	1.9910	0.0078	5.9575	0.7963	0.7917	0.7963	0.0	0.7963	0.7963
No log	2	4208	1.6121	0.0156	8.2750	0.8194	0.8176	0.8194	0.0	0.8194	0.8194
0.0708	3	6312	1.7852	0.0312	14.2331	0.8322	0.8308	0.8310	0.0	0.8322	0.8322
1.4773	4	8416	1.5433	0.0625	25.3751	0.8634	0.8634	0.8634	0.0	0.8634	0.8634
1.1626	5	10520	1.9047	0.125	47.4094	0.7986	0.7913	0.7986	0.0	0.7986	0.7986
1.0533	6	12624	1.4293	0.25	91.1283	0.8438	0.8430	0.8438	0.0	0.8438	0.8438
0.9636	7	14728	1.4806	0.5	175.1570	0.8519	0.8511	0.8519	0.0	0.8519	0.8519
0.6679	8.0	16832	1.3355	1.0	347.1295	0.8681	0.8680	0.8681	0.0	0.8681	0.8681
0.5296	9.0	18936	2.7006	1.0	342.9743	0.8519	0.8509	0.8519	0.0	0.8519	0.8519
0.4711	10.0	21040	2.2350	1.0	344.8060	0.8426	0.8422	0.8426	0.0	0.8426	0.8426
0.4328	11.0	23144	2.6491	1.0	350.0100	0.8333	0.8332	0.8322	0.0	0.8333	0.8333
0.3143	12.0	25248	3.1289	1.0	353.5838	0.8194	0.8193	0.8194	0.0	0.8183	0.8194

Framework versions

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