hlttxdy/STAR-1_DeepSeek-R1-Distill-Qwen-7B_dpo_over_refusal_mix_safep0.7_epoch2_lr1e-6_beta0.05_ftx0.2

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

safep0.7epoch2lr1e-6beta0.05ftx0.2

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-7B on the STAR-1DeepSeek-R1-Distill-Qwen-7Bdpoown and the STAR-1DeepSeek-R1-Distill-Qwen-7Bdpoownoverrefusal datasets.

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-06
• —trainbatchsize: 1
• —evalbatchsize: 8
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 4
• —gradientaccumulationsteps: 4
• —totaltrainbatch_size: 16
• —totalevalbatch_size: 32
• —optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.05
• —num_epochs: 2.0

Training results

Framework versions

• —Transformers 4.49.0
• —Pytorch 2.7.1+cu126
• —Datasets 3.2.0
• —Tokenizers 0.21.0