9a7e32a92487b8bb0030c56dc4e26fe5

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Llama-8B on the fancyzhx/dbpedia_14 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4386
• Data Size: 1.0
• Epoch Runtime: 31071.6262
• Accuracy: 0.9865
• F1 Macro: 0.9865
• Rouge1: 0.9865
• Rouge2: 0.0
• Rougel: 0.9865
• Rougelsum: 0.9865

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	23.4223	0	286.6209	0.0899	0.0456	0.0899	0.0	0.0900	0.0899
7.0511	1	17500	3.7865	0.0078	526.8343	0.6207	0.5610	0.6207	0.0	0.6206	0.6207
1.7054	2	35000	1.3672	0.0156	779.3387	0.9245	0.9244	0.9245	0.0	0.9245	0.9245
0.8982	3	52500	0.9676	0.0312	1251.2730	0.9579	0.9579	0.9579	0.0	0.9579	0.9579
0.6203	4	70000	0.8820	0.0625	2221.3219	0.9632	0.9634	0.9632	0.0	0.9632	0.9632
0.5177	5	87500	0.6354	0.125	4131.1145	0.9747	0.9748	0.9747	0.0	0.9747	0.9747
0.4422	6	105000	0.4332	0.25	7996.2066	0.9816	0.9816	0.9816	0.0	0.9816	0.9816
0.0021	7	122500	0.3079	0.5	15693.2288	0.9856	0.9856	0.9856	0.0	0.9856	0.9856
0.1966	8.0	140000	0.2828	1.0	31090.7168	0.9877	0.9877	0.9877	0.0	0.9877	0.9877
0.0755	9.0	157500	0.2940	1.0	31091.9003	0.9882	0.9882	0.9882	0.0	0.9882	0.9882
0.0648	10.0	175000	0.3568	1.0	31061.1419	0.9873	0.9873	0.9873	0.0	0.9873	0.9872
0.1227	11.0	192500	0.4238	1.0	31025.5853	0.9867	0.9867	0.9868	0.0	0.9867	0.9868
0.0484	12.0	210000	0.4386	1.0	31071.6262	0.9865	0.9865	0.9865	0.0	0.9865	0.9865

Framework versions

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