cutelemonlili/random_xoPG8CvJEE66Bcst

Distill_speculative_R1-7B

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B on the Distill_speculative_R1-7B dataset. It achieves the following results on the evaluation set:

• Loss: 0.2968

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: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 4
• total_eval_batch_size: 4
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
1.0559	0.1667	20	0.3010
0.8954	0.3333	40	0.2973
1.3178	0.5	60	0.2966
0.9733	0.6667	80	0.2962
0.7796	0.8333	100	0.2956
0.9924	1.0	120	0.2949
0.872	1.1667	140	0.2968
1.017	1.3333	160	0.2965
0.9239	1.5	180	0.2968
1.0352	1.6667	200	0.2969
0.8494	1.8333	220	0.2968
0.8062	2.0	240	0.2968

Framework versions

• Transformers 4.46.1
• Pytorch 2.6.0+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3