Models

qwen3-4b-base-svd-muon-adam-1e-6-adamlr-1e-6-bs128-kl0.0-global_step_20

qwen3-4b_grpo_skywork_math-global_step_200

qwen3_1.7b_rush_hour_one_move_4_9_epoch2

chess-special-119100

pdalma_ctx4_dm1_ce003_pr05_ptll32-1b_s2_ckpt_5_of_10_it36

pdalma_ctx4_dm1_ce0_pr1_ptll32-1b_s2_ckpt_1_of_10_it4

qwen3_1.7b_rush_hour_multi_move_final_10_12

ds1p5b_skywork_math_hard-global_step_200

qwen3_1.7b_rush_hour_multi_move_final_short_4_9_epoch2

qwen3_1.7b_rush_hour_multi_move_final_short_4_9

qwen25-3b-l3l3-ep5

k3

c67-h19

gra4

SN383

4ef9f381

nvidia_math_cot_1e5_v2_ep5

gemmasft

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

qwen3_1.7b_rush_hour_multi_move_final_short_4_9_epoch3

SDRL-rand-Qwen3-4B-Base-icml-self-debate-random_n8_l2048-DAPO_n8_bs256_long8-step200

qwen3_1.7b_rush_hour_multi_move_final_4_9_long_10_12_epoch3

qwen3-1.7b-base-instruction-tuning-full-sft

affine-5CDUswY2ZK2nXnkaWhBAWD47CQE3KvMm6AyKhJ1Txm5R5tdi

Qwen3-1.7B-CCC-merged-cp4-LR1e-4

old-122

affine-finaltest-2

Qwen3-1.7B-CCC-merged-cp5-LR1e-4

Qwen3-1.7B-CCC-merged-cp6-LR1e-4

SN381

RLAD-Sol-Gen

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