fd0be90c97567aa5774f9d0009941641

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B on the fancyzhx/dbpedia_14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2764
• Data Size: 1.0
• Epoch Runtime: 3781.8311
• Accuracy: 0.9897
• F1 Macro: 0.9897
• Rouge1: 0.9897
• Rouge2: 0.0
• Rougel: 0.9897
• Rougelsum: 0.9898

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	14.7386	0	131.6403	0.0847	0.0473	0.0848	0.0	0.0847	0.0847
0.8489	1	17500	0.3926	0.0078	160.9834	0.9787	0.9787	0.9787	0.0	0.9787	0.9788
0.3574	2	35000	0.3436	0.0156	188.6564	0.9823	0.9824	0.9823	0.0	0.9823	0.9824
0.3144	3	52500	0.4224	0.0312	247.3292	0.9817	0.9816	0.9817	0.0	0.9816	0.9817
0.2954	4	70000	0.2927	0.0625	362.8408	0.9848	0.9848	0.9848	0.0	0.9848	0.9848
0.2063	5	87500	0.2561	0.125	588.1284	0.9874	0.9874	0.9875	0.0	0.9874	0.9874
0.2603	6	105000	0.2110	0.25	1036.8979	0.9876	0.9876	0.9876	0.0	0.9876	0.9876
0.0014	7	122500	0.2093	0.5	1950.8366	0.9878	0.9878	0.9878	0.0	0.9878	0.9878
0.1219	8.0	140000	0.1926	1.0	3777.7658	0.9899	0.9899	0.9899	0.0	0.9899	0.9899
0.0937	9.0	157500	0.2016	1.0	3785.8869	0.9902	0.9902	0.9902	0.0	0.9902	0.9902
0.0692	10.0	175000	0.2419	1.0	3796.6584	0.9898	0.9898	0.9898	0.0	0.9898	0.9898
0.053	11.0	192500	0.2823	1.0	3817.2150	0.9897	0.9897	0.9897	0.0	0.9896	0.9897
0.0693	12.0	210000	0.2764	1.0	3781.8311	0.9897	0.9897	0.9897	0.0	0.9897	0.9898

Framework versions

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