Lanxiang Hu's picture

Lanxiang Hu PRO

Snyhlxde

·

https://snyhlxde1.github.io/

snyhlxde1

AI & ML interests

Machine learning systems, distributed learning, efficient machine learning on edge devices

Recent Activity

liked a model 15 days ago

GEAR-Dreams/DreamZero-DROID

liked a dataset about 2 months ago

lmgame/VideoScienceBench

upvoted an article 2 months ago

Forge: Scalable Agent RL Framework and Algorithm

View all activity

Organizations

Snyhlxde 's models 72

Snyhlxde/openthought2_mathsubset

Updated Sep 9, 2025

Snyhlxde/shiftedattn-9-3-coder-7B-ntok16_soft_ce_oci_datav1_59k_stp_ar_10_cyclic_prog_noise_all_lr1e-6

8B • Updated Sep 4, 2025 • 1

Snyhlxde/shiftedattn-9-2-coder-7B-ntok16_soft_ce_oci_datav1_59k_stp_ar_10_cyclic_prog_noise_all_lr2e-6

8B • Updated Sep 3, 2025 • 1

Snyhlxde/shiftedattn-8-31-coder-7B-ntok16_soft_ce_oci_datav1_59k_stp_ar_10_cyclic_prog_noise_all_lr5e-6

8B • Updated Sep 2, 2025 • 1

Snyhlxde/shiftedattn-8-31-coder-7B-ntok16_soft_ce_oci_datav1_44p5k_stp_ar_10_cyclic_prog_noise_all_lr5e-6

8B • Updated Sep 1, 2025 • 3

Snyhlxde/shiftedattn-8-28-coder-7B-ntok16_soft_ce_oci_datav1_40k_smpl_ar_10_cyclic_prog_noise_r0p5_lr5e-6

8B • Updated Sep 1, 2025 • 1

Snyhlxde/8-28-qwen25-coder-7B-ntok16_soft_ce_flexattn_oci_data_v1_40k_smpl_ar_10_cyc_prog_noise_lr5e-6

8B • Updated Aug 29, 2025 • 1

Snyhlxde/logitsaligned-8-27-qwen2p5-coder-7B-ntok16_soft_ce_flexattn_oci_data_v1_40k_smpl_ar_10_lr5e-6

8B • Updated Aug 28, 2025 • 1

Snyhlxde/logitsaligned-8-27-qwen2p5-coder-7B-ntok16_soft_ce_flexattn_oci_data_v1_10k_smpl_ar_10_lr5e-6

8B • Updated Aug 28, 2025 • 1

Snyhlxde/alignedlogits-8-24-7B-ntok64_ce_soft_loss_length_20k_flexattn_32kdata_v1_ar_10_lr5e-6

8B • Updated Aug 25, 2025 • 1

Snyhlxde/alignedlogits-8-24-7B-ntok64_ce_soft_loss_length_20k_flexattn_32kdata_v1_ar_1_only_lr5e-6

8B • Updated Aug 25, 2025 • 1

Snyhlxde/aligned-8-24-7B-ntok64_ce_soft_loss_length_20k_flexattn_data_v1_4k_smp_ar_10_lr5-6

8B • Updated Aug 24, 2025 • 1

Snyhlxde/aligned-8-24-7B-ntok64_ce_soft_loss_length_20k_flexattn_data_v1_4k_smp_ar_1_only_lr5-6

8B • Updated Aug 24, 2025 • 2

Snyhlxde/8-23-openthinker2-7B-ntok64_ce_soft_loss_length_capped_16k_flexattn_data_v1_4kexample_ar_10

8B • Updated Aug 24, 2025 • 1

Snyhlxde/8-23-openthinker2-7B-ntok64_ce_soft_loss_length_capped_16k_flexattn_data_v1_4kexample_ar_only

8B • Updated Aug 24, 2025 • 2

Snyhlxde/openthinker2-7B-ntok64_soft_ce_loss_length_20k_flexattn_data_v1_32k_sample_ar_10_c_d

8B • Updated Aug 22, 2025 • 2

Snyhlxde/openthinker2-7B-ntok64_soft_ce_loss_length_20k_flexattn_data_v1_4k_sample_ar_10_c_d

8B • Updated Aug 21, 2025 • 1

Snyhlxde/soft_loss_length_capped_16k_flexattn_data_v1_4k4samples_8_16_no_prompt_boundary

8B • Updated Aug 17, 2025 • 1

Snyhlxde/openthoughts13k-keep_ratio0.15_soft_cllm_loss_downsampling_training

8B • Updated Aug 5, 2025 • 2

Snyhlxde/without_think_format_tetris_3by3_0514

Updated May 15, 2025

Snyhlxde/with_think_format_tetris_3by3_0514

Updated May 14, 2025

Snyhlxde/without_think_format_tetris_0513

Updated May 14, 2025

Snyhlxde/with_think_format_tetris_0513

Updated May 14, 2025

Snyhlxde/with_think_format_tetris_0512

Updated May 12, 2025

Snyhlxde/without_think_format_tetris_0511

Updated May 12, 2025

Snyhlxde/deprecated_tetris_dev

Updated May 11, 2025

Snyhlxde/DeepSeek-R1-Distill-Llama-8B-openthought-vllm

Updated Apr 12, 2025 • 1

Snyhlxde/base_207M_v1_decay0p99_2

Updated Feb 13, 2025

Snyhlxde/base_207M_v1_fliprob0.03_decay1p0

Updated Feb 12, 2025

Snyhlxde/base_207M_v1_decay0p99

Updated Feb 12, 2025