4-5B Parameter LLMs — Page 36

qwen3_4b_sudoku_multi_act_rl

qwen3_4b_sudoku_multi_act_rl_allow_one_action_epoch1

qwen3_4b_sudoku_multi_act_rl_allow_one_action_epoch2

qwen3_4b_sudoku_one_act_rl_default_epoch2

qwen3_4b_sudoku_one_act_rl_default_epoch3

qwen3_4b_sudoku_multi_act_rl_allow_one_action_epoch3

4b_sft_ds_rea_epoch3

c21

gemma3-4b-mbti-chat-energy

Diabetica-Qwen3-4B

Qwen3-4B-Thinking-2507-GLM-4.6-Distill

online_acemath_rl_4b_inst_hard_16k_thinking_no_summ_thinking_step_90

Qwen3-4B-rft-alfworld

Affine-sniper_v5

Affine-dyna2

affine-tobetop1

multiturn-sft-qwen-3-4b

Affine-silly-mistake

qwen3-4b-thinking-rl-ckpt60

Affine-pipi_v1

Affine-odor_v4

qwen3_4b_easy_rl_new

Affine-taichi38

SkeptiSTEM-4B-stageR1-merged-16bit

Affine-1912-1936

Affine-v1

qwen3_4b_easy_rl_our_adv_final

affine-legacy

Affine-h01

Affine-h02

Qwen3-4B-Tulu-SFT

Affine-0101928-rise

affine-mighty-eagle-999999

affine-1

qwen3-4b-elicit-pos-ckpt72

Qwen3-4B-Instruct-2507-Hanabi-SFT

self-debate-exp-Qwen3-4B-Base-majority_n4_l2048-DAPO_n8_bs256_long8-step200

dyck-test

Affine-5Dc4pnGJtH93eRjpuZoF1KnvxvkEFQV5LZiuP1RJjfMinxt4

Qwen3-4B-rft-alfworld-e5

stockr1-qwen3-4b

affine-bug-5E7XUcHcvGaeU2jRXPLPdpwPy6D3dF55Ujpiy3VwN9TE4A5f