Models

ds-svd-muon-adam-1e-6-global_step_140

ds-adam-1e-6-global_step_40

ds-adam-1e-6-global_step_100

ds-adam-2e-6-global_step_200

ds1p5b_skywork_math_hard-global_step_200

ds1p5b_skywork_math_hard-global_step_400

sft_qwen15_code200_lr_1e-5_cosine_bsz_128_ckpt_2_of_5

sft_qwen15_code200_lr_5e-6_constant_bsz_128_ckpt_1_of_5

sft_qwen15_code200_lr_5e-6_constant_bsz_64_ckpt_3_of_5

pdcd200_cptq15_ce003_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_2_of_10_it26

pdcd200_cptq15_ce003_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_1_of_10_it15

pdcd200_cptq15_ce003_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_3_of_10_it45

pdcd200_cptq15_ce003_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_7_of_10_it387

pdcd200_cptq15_ce003_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_3_of_10_it45

pdcd200_cptq15_ce003_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_5_of_10_it132

pdcd200_cptq15_ce003_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_7_of_10_it387

pdcd200_cptq15_ce003_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_9_of_10_it1135

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_3_of_10_it45

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_4_of_10_it77

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_6_of_10_it226

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_7_of_10_it387

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_8_of_10_it663

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_10_of_10_it1943

pdcd200_cptq15_ce01_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_1_of_10_it15

pdcd200_cptq15_ce01_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_2_of_10_it26

pdcd200_cptq15_ce01_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_4_of_10_it77

pdcd200_cptq15_ce01_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_5_of_10_it132

pdcd200_cptq15_ce01_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_1_of_10_it15

pdcd200_cptq15_ce01_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_3_of_10_it45

pdcd200_cptq15_ce01_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_4_of_10_it77

pdcd200_cptq15_ce01_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_10_of_10_it1943

hot_start_1.7b_bt_oracle_kl_1e-3_770

pdcd200_cptq15_ce003_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_9_of_10_it1135

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_1_of_10_it15

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_2_of_10_it26

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_3_of_10_it45

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_4_of_10_it77

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_8_of_10_it663

pdcd200_cptq15_ce01_pr1_ptq25-15b_omi_c100k_200tok_s8_ckpt_9_of_10_it1135

pdcd200_cptq15_ce01_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_9_of_10_it1135

pdcd200_cptq15_ce01_pr05_ptq25-15b_omi_c100k_200tok_s8_ckpt_10_of_10_it1943

pdcd200_cptq15_ce0_pr0_ptq25-15b_omi_c100k_200tok_s8_ckpt_1_of_10_it15