16b_RL
Vanilla_RL
qwen3_1.7b_easy_rl_ours_adv_fixed_gamma_1_98_gem_ms_seq_is
Advanced_Risk_Summarization_Qwen3-4B-Base
Qwen3-0.6B-Reverse-Text-SFT
64b_SFT
qwen3_1.7b_easy_rl_ours_adv_fixed_gamma_1_98_geo_ms_token_tis
qwen3_1.7b_easy_rl_ours_adv_fixed_geo_ms_token_tis
qwen3_1.7b_easy_rl_ours_adv_fixed_geo_ms_seq_is_epoch3
qwen3_1.7b_easy_rl_ours_adv_fixed_geo_ms_seq_is
qwen3_1.7b_new_standard_A_sft_overfit_lr_5e_6__global_step_96
qwen3_1.7b_new_standard_A_sft_overfit_lr_5e_6__global_step_192
qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_594
qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_198
qwen3_1.7b_new_standard_B_sft_overfit_lr_5e_6__global_step_792
qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_1480
qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_1776
qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_888
qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_5__global_step_592
qwen3_1.7b_new_standard_A_sft_overfit_lr_5e_6__global_step_288
qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_1184
qwen3_1.7b_new_standard_C_sft_overfit_lr_5e_6__global_step_296
qwen3_1.7b_rush_hour_one_move_sft
qwen317step300
64b_RL_DAPO
32b_RL_DAPO
1b_RL_DAPO
qwen3_1.7b_sudoku_multi_action_sft_final
run0118-local-reasoning-obo-0_5-baseline-max32-step49
affine-v-9-5EWSasAgABTaNwkLMudKKCZw8WZKbiNMcQrHKUUMwMoWsxRj
qwen3_1.7b_rush_hour_one_move_final
Affine-5Dc4pnGJtH93eRjpuZoF1KnvxvkEFQV5LZiuP1RJjfMinxt4
2b_SFT
qwen3_1.7b_new_sudoku_one_action_B_sft_lr_5e_6__step_1108
qwen3-1.7b-base-adam-3e-6-bs128-kl0.0-global_step_200
Affine-5GYdM3kPgYkco7VwEvG356Si6xkk1Ae4iurBJ6YGf7vTAFuX
qwen3_1.7b_sudoku_multi_action_easy_11_20
qwen3_1.7b_new_sudoku_one_action_B_sft_lr_5e_6__step_3324
qwen3_1.7b_new_sudoku_one_action_A_sft_lr_5e_6__step_562
Affine-at01-12-31-01
Affine-at02-12-31-02
merge_lenfmt_MRL4096_ROLLOUT4_LR5e-7_w0.5_tall_mask_ta
