bb803c19aabd4996b1cdd9983bb042b2

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B on the google/boolq dataset. It achieves the following results on the evaluation set:

• Loss: 2.7283
• Data Size: 1.0
• Epoch Runtime: 76.8330
• Accuracy: 0.7849
• F1 Macro: 0.7556
• Rouge1: 0.7849
• Rouge2: 0.0
• Rougel: 0.7846
• Rougelsum: 0.7852

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	7.7282	0	7.3503	0.3922	0.3140	0.3925	0.0	0.3925	0.3925
No log	1	294	9.0113	0.0078	8.3528	0.3934	0.3108	0.3937	0.0	0.3935	0.3934
No log	2	588	4.3365	0.0156	9.2905	0.6161	0.4068	0.6160	0.0	0.6155	0.6161
No log	3	882	2.4873	0.0312	12.1349	0.6492	0.5399	0.6489	0.0	0.6489	0.6489
0.1456	4	1176	2.6503	0.0625	15.2500	0.6838	0.5702	0.6841	0.0	0.6835	0.6841
0.1775	5	1470	2.0410	0.125	19.8775	0.7644	0.7339	0.7647	0.0	0.7641	0.7650
0.2947	6	1764	2.5183	0.25	27.8321	0.6562	0.6556	0.6562	0.0	0.6566	0.6559
1.438	7	2058	2.1869	0.5	44.5707	0.7463	0.7432	0.7460	0.0	0.7457	0.7463
1.1844	8.0	2352	2.1327	1.0	79.8337	0.8079	0.7948	0.8079	0.0	0.8076	0.8076
0.6149	9.0	2646	2.7283	1.0	76.8330	0.7849	0.7556	0.7849	0.0	0.7846	0.7852

Framework versions

• Transformers 4.57.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.1